claude-plugins-official

The Ralph Wiggum Loop Explained: Why AI Agents Quit Long Tasks Halfway (and the Simple Fix) This guide was first published on saurabhbhayana.com. Imagine asking an AI agent to rebuild your 30-page website overnight. In the morning the homepage looks great. Then you click around. The contact form goes nowhere, two pages are blank, and the agent's last message says every feature is complete and tested. Annoying, right? Here is the funny part. The fix that went viral among developers is named after Ralph Wiggum, the not-very-bright kid from The Simpsons. In its simplest form it fits on one line. And the idea behind it works even if you never write a line of code. In this post we explain why AI agents give up on long tasks, what the Ralph Wiggum loop is, why it works, how to run it in Claude Code, and how a business owner can use the same trick in a normal ChatGPT or Claude chat. The starting point was a good explainer on agent harnesses, which we link at the end. If the words prompt, context and harness engineering are new to you, our plain-English guide to those terms is a good first read. Why do AI agents stop halfway on long tasks? Every AI model has a context window: the text it can hold in mind at once. Your instructions, the files it read, the code it wrote and every message so far all sit in there. When ChatGPT launched in 2022 that window was about 4,000 tokens, a few pages. Today's windows are far bigger, so agents now take on jobs that run for hours. But a 12-hour job still fills any window. So what happens when it fills? Most agents compact. They squeeze the conversation into a summary and keep going from the summary. A lot of people think that solves the problem. Not really. A summary is only as good as the agent's guess about what mattered. Details drop out, like which button was never wired up, and sometimes the summary says a feature is done when it was only started. > An agent working from its own summary is like a contractor working from a sticky note that says "kitchen: done". Anthropic saw the same failures when it pushed its own models on long builds. In a November 2025 engineering post it wrote that compaction isn't sufficient, and named two habits: the agent tries to build everything at once and runs out of room halfway through a feature, and a later session looks at the progress so far and declares the job done. So if more memory and better summaries don't fix it, what does? What is the Ralph Wiggum loop? In July 2025 developer Geoffrey Huntley published a technique he called Ralph. In its purest form it is a single line of bash: while :; do cat PROMPT.md | claude-code ; done Read it out loud and it says: forever, feed the same prompt file to the coding agent. When the agent finishes, start it again. Same instructions, brand new session, empty context window. That's it. The trick is in what the prompt tells the agent to do on every pass. Huntley's rules are simple: read the spec, look at the plan file, choose the most important thing that is not done, do that one item, run the tests for it, update the plan, and stop. Then the loop starts a fresh agent, which reads the updated plan and picks the next item. !The Ralph Wiggum loop: a fresh agent starts, reads the spec and plan files, picks one task, does it, tests it, writes progress back to the files, exits, and the loop starts a fresh agent again until every task passes The memory lives in the files. Every pass starts with an empty head and a clear note. Picture a building site that runs in shifts. Nobody expects the night crew to remember what the day crew did. They read the handover sheet, do their part, update the sheet and go home. The Ralph loop gives an AI agent a handover sheet instead of asking it to remember everything. Ralph Wiggum loop: why is it called that? Huntley's original post doesn't spell it out, but the joke is easy to read. Ralph Wiggum is cheerful, not very bright and never gives up. The README for the official Claude Code plugin describes it as persistent iteration despite setbacks. Huntley himself calls the technique deterministically bad in an undeterministic world: each pass may be a bit clumsy, but you know exactly what it will try, and enough passes get the job done. There is a second lesson in the name. You don't need a genius agent for long work. You need an ordinary one with good notes, small steps and a test at the end of each step. Which raises the question: what exactly goes in those notes? How the Ralph loop works, step by step 1. Write the spec with the agent. Talk through what you want built until it is written down as a requirements file. Huntley does this as a conversation at the start of the project. 2. Turn the spec into a task list. Every feature becomes one line or one entry, and every entry starts as not done. Huntley calls his file fix_plan.md. Anthropic uses a JSON file where each feature has a passes field set to false. 3. Loop: one task per pass. Each fresh agent reads the spec and the list, picks the most important task that is not done, and works only on that. 4. Test before you tick. The agent runs the tests for that task. Huntley calls this backpressure: the tests push back when the work is wrong, so the agent can't just claim success. 5. Write it down. The agent updates the task list and a progress note, and in code projects commits the change to git, so the next pass knows exactly where things stand. 6. Exit and repeat. The agent stops, the loop starts a fresh one, and this continues until every task passes or you hit your limit. | | One long session | Ralph loop | | --- | --- | --- | | Memory | The chat, squeezed into summaries | Files: spec, task list, progress note | | Scope per step | Tries to do everything at once | One task per pass | | When it says done | Its own judgment | The tests and the task list | | If a step goes wrong | The mistake stays in context | Next pass starts clean; git can undo it | | Best for | Jobs that fit in one sitting | Jobs with many separate pieces | !One long agent session versus the Ralph loop: the long session fills its context window, compacts into a summary, loses details and declares the job done early, while the Ralph loop starts fresh each pass and reads the task list and progress file Same model, same job. The difference is where the memory lives. What Anthropic does for long-running agents Anthropic's engineering guide to long-running agents is worth reading even if you never touch the code, because it is the same idea done carefully. The first session is an initializer agent with its own prompt. It sets up the project and writes a feature list. For their test project, a clone of the claude.ai app, that list had more than 200 features, all marked failing. Every session after that is a coding agent with a fixed routine: read the progress file, called claude-progress.txt, and the git history, run a start-up script, check that the app still works end to end, then build one feature, test it in a real browser as a user would, commit it and update the progress file. The agent may only flip a feature's status. The guide is blunt about it: it is unacceptable to remove or edit tests. > Tip: Steal the testing rule even for non-code work. Write the check before the work starts, such as "page loads on a phone and the form sends an email", and never let the AI change the check to make itself pass. How to use the Ralph loop in Claude Code Anthropic's official plugin marketplace has a Ralph Loop plugin for Claude Code. Once it is installed, you start a loop with one command: /ralph-loop "Build the booking page from SPEC.md. Work through TASKS.md one item at a time, run the tests, update TASKS.md. Output <promise>COMPLETE</promise> when every task passes." --completion-promise "COMPLETE" --max-iterations 30 Claude works on the task and tries to finish. A Stop hook blocks the exit and feeds the same prompt back in. That repeats until Claude prints your completion phrase or the pass limit is reached. To stop it early, run /cancel-ralph. One honest detail most guides skip: the plugin loops inside your current session. Claude sees its own files and git history each time, which is the core of Ralph, but the conversation keeps growing rather than starting empty. Huntley's original bash loop starts a brand new agent every pass. For very long jobs, we prefer the original shape, or at least a task list and progress file that the prompt tells Claude to read first. The same bash loop works with other command-line coding agents too. You only need an agent that can take a prompt from a file and exit when it is done. Use the Ralph idea without writing code You don't need a terminal to borrow this. Say you want AI to write 40 product descriptions, or turn a 25-page website into new copy. Doing it in one giant chat is the long session from the table above: by item 30 the answers get vague, and the AI forgets the rules you gave it at the start. Do it the Ralph way instead: - Keep a master document with the brief and your rules at the top, and a task list below it with every item marked not done. - Start a new chat for each item, or a small batch. Paste in the brief, the rules and the progress so far. - Ask for one item only, and check it against your rules before you accept it. - Mark it done in the master document and add a one-line note on anything you changed. - Repeat. If one chat goes off the rails, you lose one item, not the whole job. !The Ralph idea without code: a master document with the brief and task list, a new chat per item, one item at a time checked against rules, then mark it done and repeat The no-code version. A new chat is the cheapest fresh context window there is. This is the same pattern we use for big content and website jobs for clients. It is also why projects in ChatGPT and Claude are so useful: the brief and rules live in the project, and each chat starts clean. When you don't need a Ralph loop Be honest about the size of the job. If a task fits comfortably in one session, say a single page or a short script, a loop just adds setup. Ralph pays off when the work splits into many separate pieces that can each be checked. Watch the bill too. Every pass reads the files again and costs tokens, so a loop with no limit can keep spending long after the useful work is done. Always set a maximum number of passes, and don't leave an agent looping unattended on anything live, like a production website or a customer database. Give it a copy to work on. The bigger picture Strip away the funny name and the lesson is old: for long work, memory belongs on paper, not in someone's head. Factories run on shift handovers. Hospitals run on patient charts. AI agents turn out to work the same way. The model matters, but the notes, the small steps and the tests matter more than most people expect. So, what is the long job you keep putting off because the AI loses the plot halfway? Write the task list first, and see how far an ordinary agent gets with good notes. If you want help setting up a loop for a real project, tell us what you are building. Sources: the YouTube explainer on agent harnesses that inspired this post, at youtube.com; Geoffrey Huntley, "Ralph Wiggum as a software engineer", July 2025; Anthropic Engineering, "Effective harnesses for long-running agents", November 2025; and the Ralph Loop plugin README in Anthropic's official plugin repository. Takeaways - AI agents quit long tasks because their working memory, the context window, fills up. Squeezing it into a summary loses detail, and the agent can wrongly decide the job is finished. - The Ralph Wiggum loop, published by Geoffrey Huntley in July 2025, runs the agent again and again with the same prompt. Each pass picks one task from a plan file, does it, tests it, writes down what it did, and exits. - The memory lives in files, not in the chat: a spec, a task list, a progress note and the work itself. A fresh pass reads them and carries on. - Anthropic's own guide to long-running agents uses the same shape: a feature list marked failing, a progress file, git commits and one feature at a time. - Claude Code has an official Ralph Loop plugin: /ralph-loop with a maximum number of passes and a completion phrase. It loops inside the same session, unlike the original, which starts the agent fresh each time. - You can use the idea without code: one master task list, one task per chat, and a progress note you paste in at the start and update at the end. - Always cap the number of passes. Every pass costs tokens, and a loop with no limit can run up a bill. FAQs What is the Ralph Wiggum loop? A technique published by Geoffrey Huntley in July 2025 that runs an AI coding agent in a loop with the same prompt. Each pass reads a spec and a task list, does one task, tests it, records progress in files and exits, and the next pass starts fresh. The memory lives in the files instead of the chat. Why is it called the Ralph Wiggum loop? It is named after Ralph Wiggum from The Simpsons. The official Claude Code plugin describes the idea as persistent iteration despite setbacks: an ordinary agent that keeps trying, one small step at a time, gets long jobs done. How do I use the Ralph loop in Claude Code? Install the Ralph Loop plugin from Anthropic's official plugin marketplace, then run /ralph-loop with your prompt, a --completion-promise phrase and a --max-iterations limit. Use /cancel-ralph to stop it early. Why do AI agents stop halfway on long tasks? Their context window fills up. Most agents then compact the conversation into a summary, which loses details and can wrongly record work as finished, so the agent stops early or leaves features half built. What is context compaction? When an AI agent's context window is nearly full, it summarises the conversation so far and continues from the summary. It lets the agent keep working, but details are lost, which is why long tasks can go wrong after it happens. Does the Ralph loop work with Cursor, Codex or other agents? The original idea is a shell loop that feeds the same prompt file to a command-line agent, so it can work with any coding agent you can start from a prompt and that exits when it is done. The task list and progress file matter more than the tool. Is the Ralph loop expensive? It can be. Every pass reads the files again and uses tokens, so always set a maximum number of passes and check progress before letting it run longer. --- Originally published at saurabhbhayana.com, where I write about AI automation, AI search and Local SEO for small businesses.

Claude Code GitHub Actions: Setup, @claude Mentions and Automatic PR Review Claude Code GitHub Actions: Setup, @claude Mentions and Automatic PR Review The first time you tag @claude on a GitHub issue and watch a pull request show up twenty minutes later, it feels a little like cheating. You wrote two sentences. Something read the repo, made a branch, ran the checks and left a comment explaining what it did. Then the second feeling arrives. Who exactly is allowed to trigger that? What is it allowed to touch? And what does each of those twenty minutes cost? This guide covers the Claude Code GitHub integration end to end: the one-command setup, the manual setup, the @claude workflow, an automatic pull request review workflow, the difference between the open-source action and the managed Code Review product, how to cap the bill, and the security settings I would not skip. I checked it against Anthropic's documentation on October 29, 2026. The action and the app change fast, so where this page and the docs disagree, trust the docs. Sources are at the bottom. What "Claude Code GitHub" Actually Means Claude Code GitHub Actions is a GitHub Action, anthropics/claude-code-action, that runs Claude Code inside your repository's workflows. It responds to @claude mentions and can also run a prompt automatically on any GitHub event. The confusing part is that several things share the Claude Code name on GitHub. The docs say this outright, and it is worth getting straight before you configure anything: - The GitHub Action (this guide). You add workflow files to the repo, and the job runs on your GitHub runners with your credentials. - Code Review. A managed service that reviews every pull request on Anthropic infrastructure. No workflow file. It is for Team and Enterprise plans. - The /code-review command. You run it inside a local Claude Code session on your own branch. Any plan. - Claude Code in the cloud. Sessions that run on cloud infrastructure instead of your laptop, including auto-fix for pull requests. - The Agent SDK. The action is built on it. You use it directly when you want automation outside GitHub. If your goal is "tag Claude in an issue and get a PR", you want the action. If your goal is "every PR gets a review without anyone maintaining YAML", look at Code Review, which I cover further down. If you only want a review of your own branch before you push, the local command is enough. Everything below is the action unless I say otherwise. Quick Setup With /install-github-app The fastest path is one slash command. You need admin access to the repository, and the GitHub CLI installed and authenticated with gh auth login. Claude Code checks for gh and warns you if it is missing. cd your-repo claude Then inside the session: /install-github-app Follow the prompts. Claude Code does four things for you: 1. Installs the Claude GitHub App on the repository. 2. Sets up an authentication secret. If Claude Code already has an API key it reuses it and offers to keep an existing ANTHROPIC_API_KEY secret. Otherwise you pick between creating a long-lived token tied to your Claude subscription or pasting an API key. 3. Saves the credential as a repository secret, named ANTHROPIC_API_KEY for a key or CLAUDE_CODE_OAUTH_TOKEN for a subscription token. 4. Pushes a branch with the workflow files you selected, already wired to that secret, and opens GitHub in your browser with a pull request ready to create. Merge that pull request and @claude works in the repo. That is the whole setup. A few details from the docs that save you time: - github.com only. If your remote is on GitLab or Bitbucket, the command prints a notice and exits. GitLab has its own CI/CD integration. - You can bail out. Press Esc to stop partway. A step already running finishes, nothing later starts, and the closing message lists what already happened, such as a pushed branch or a saved secret. - You can install the app now and finish later. Choose "Skip for now" after the app install, then run /install-github-app again when you are ready for the workflow and secret. - It can update old workflows. If the repo already has a claude.yml, rerun the command and choose "Update workflow file with latest version". It opens a fresh PR with the new files. - The review workflow is optional. If you select it during setup, Claude posts each review on the PR as inline comments, or one summary comment when it finds nothing. Before v2.1.229 it wrote reviews only to the workflow run log, so if your review workflow was generated earlier, update it. Quick setup works with the Claude API and Claude subscriptions. If you run Claude Code through Amazon Bedrock, Google Cloud's Agent Platform or Microsoft Foundry, the setup is different and covered in the cloud providers section below. Manual Setup: App, Secret, Workflow File Use the manual path when you do not run Claude Code locally, when the command fails, or when you want full control of the files. 1. Install the Claude GitHub App. Install it on the repository. The action relies on three of its permissions: Contents (read and write), Issues (read and write) and Pull requests (read and write). 2. Add a secret. Add one of these as a repository secret: - ANTHROPIC_API_KEY: a key from the Claude Console. - CLAUDE_CODE_OAUTH_TOKEN: a token that authenticates with your subscription, available on Pro, Max, Team and Enterprise. Generate it by running claude setup-token locally. In workflow files, the API key goes to the anthropic_api_key input and the token goes to claude_code_oauth_token. 3. Copy the workflow. Take examples/claude.yml from the action's repository into .github/workflows/. As committed it responds to @claude mentions using ANTHROPIC_API_KEY. If you use the OAuth token instead, change that one line to claude_code_oauth_token: ${{ secrets.CLAUDE_CODE_OAUTH_TOKEN }}. Test it by tagging @claude in an issue or PR comment. API Key or Subscription Token? This is the choice people get wrong, so here is how I would decide: - Solo repo, personal project: the subscription token is fine. Runs use your subscription instead of API billing, which means they also count against your allowance. If you have ever hit the wall, read [how Claude Code usage limits work](/blog/claude-code-usage-limits-2026/) before you let a bot spend it for you on every PR. - Team repo or organization: use an API key. The docs make the point directly: an OAuth token is tied to the subscription of the person who ran claude setup-token. When that person leaves or changes plan, every repo using their token breaks, and their personal allowance is what is paying for the team's reviews. - No long-lived secret at all: use workload identity federation. The action exchanges the workflow's GitHub OIDC token for API access through a Claude Console service account. You set anthropic_federation_rule_id and anthropic_organization_id, optionally the service account and workspace IDs, and grant the job id-token: write. The @claude Workflow, Line by Line Here is the workflow from the docs for responding to mentions: name: Claude Code on: issue_comment: types: [created] pull_request_review_comment: types: [created] jobs: claude: if: contains(github.event.comment.body, '@claude') runs-on: ubuntu-latest permissions: contents: write pull-requests: write issues: write id-token: write actions: read steps: - uses: actions/checkout@v6 with: fetch-depth: 1 - uses: anthropics/claude-code-action@v1 with: anthropic_api_key: ${{ secrets.ANTHROPIC_API_KEY }} The parts that are not boilerplate: - id-token: write is required for the action's default GitHub App authentication. - actions: read lets Claude read CI results on pull requests. That is how it can say "the tests failed on this line" instead of guessing. - actions/checkout gives Claude a local copy of the repo to work in. - if: contains(...) keeps runners from starting on comments that do not mention @claude. The action also checks the trigger phrase itself, but the if saves you a runner spin-up on every unrelated comment. Once it is merged, you talk to it in plain language: @claude implement this feature based on the issue description @claude how should I implement user authentication for this endpoint? @claude fix the TypeError in the user dashboard component Claude replies in a comment on the same issue or PR and updates that comment as it works. The trigger phrase is @claude by default and can be changed with trigger_phrase. One thing the docs state that is easy to miss: the comment must contain @claude as a complete word. /claude and @claude-bot do not count. What Makes a Good @claude Request The action is only as good as the instruction. The cost section below says the same thing from the money side: specific requests need fewer turns. Three habits that help: - Point at files and symptoms. "Fix the TypeError in UserDashboard.tsx when user.profile is null" beats "fix the dashboard bug". - Say what done looks like. "Add a test that fails before the fix" gives it a finish line. - Ask questions first when the design is open. @claude how would you approach this? costs less than a branch you throw away. It is the same logic as [plan mode](/blog/claude-code-plan-mode-2026/): decide before you build. Interactive Mode vs Automation Mode The action picks its mode from your workflow, with no mode input to set. - Interactive mode. No prompt input. Claude waits for the trigger phrase in an issue comment, a PR comment, a PR review, or the title or body of a newly opened issue, then responds. Progress and results show up as a comment. - Automation mode. You provide a prompt. Claude runs on whatever event triggers the workflow, without anyone mentioning it. By default the output goes to the workflow run log, not a comment. It can post to the issue or PR only when your prompt tells it to and it has a tool that can post. That second point trips people up. You set up an automatic reviewer, the job goes green, and nothing appears on the pull request. The review is in the run log. Posting is opt-in, and the next section shows how. Automatic Pull Request Review This is the workflow most teams actually want. It runs on every pull request and posts a review. The docs example uses the official code-review plugin: name: Code Review on: pull_request: types: [opened, synchronize, ready_for_review, reopened] jobs: review: runs-on: ubuntu-latest permissions: contents: read pull-requests: read issues: read id-token: write steps: - uses: actions/checkout@v6 with: fetch-depth: 1 - uses: anthropics/claude-code-action@v1 with: anthropic_api_key: ${{ secrets.ANTHROPIC_API_KEY }} plugin_marketplaces: "https://github.com/anthropics/claude-plugins-official.git" plugins: "code-review@claude-plugins-official" prompt: "/code-review:code-review --comment ${{ github.repository }}/pull/${{ github.event.pull_request.number }}" claude_args: '--allowedTools "mcp__github_inline_comment__create_inline_comment"' Two lines decide where the review goes: - --comment makes Claude post the review on the PR, as an inline comment on each issue or one summary comment when it finds none. Leave it off and nothing is posted. - claude_args with --allowedTools has to stay, even though the skill names the same tool in its own frontmatter. The action only starts the MCP server that posts inline comments when --allowedTools in claude_args names it. Notice the permissions. This job gets contents: read and pull-requests: read, not write. A reviewer does not need to push code. Giving it less is free and removes a whole class of surprises. What it skips, per the docs: draft and closed pull requests, pull requests it judges not to need a review (automated or trivial ones), and pull requests that already have a comment from Claude. Two practical limits: - Fork PRs on public repos. GitHub withholds secrets from runs triggered by fork pull requests, so this review only runs on PRs from branches in the same repository. - It is not a merge gate. Treat the output as a second reader, not an approver. My own take is that it should never be the only review on code that touches auth, billing or data deletion. The [pull request manifest](/blog/ai-agent-pull-request-review-policy-2026/) I wrote about earlier is a better way to make agent-written PRs checkable by a human, whichever tool wrote them. You are not locked into the plugin either. The prompt input accepts plain text or a skill invocation. For a skill that lives in your repo's .claude/skills/ directory, run actions/checkout first and pass /skill-name as the prompt. If you want to write your own review skill with your team's rules in it, [writing a Claude Code skill](/blog/writing-claude-code-skill-developer-guide-2026/) walks through the format. Claude Code Action vs Managed Code Review There are two ways to get automatic reviews on GitHub, and they are different products. This table is the short version: | | GitHub Action (this guide) | Managed Code Review | | --- | --- | --- | | Setup | Workflow files in your repo | An Owner turns it on in admin settings | | Runs on | Your GitHub runners | Anthropic infrastructure | | Plans | Anyone with an API key or cloud credentials | Team and Enterprise | | Billing | Actions minutes plus API tokens or subscription | Managed, with a monthly spend limit | | Customization | Your prompt, plugins, CLAUDE.md, any trigger | REVIEW.md and CLAUDE.md | | Beyond review | Edits code, pushes commits, runs any prompt | Review only, read-only access, never pushes | How the managed product works, per the docs: several agents read each change alongside the surrounding code, a verification step checks the findings against the code to filter false positives, the rest are deduplicated and ranked by severity, and the results land as inline comments, a summary and a check run called Claude Code Review. The check run does not conclude as a failure, and a review never approves or blocks the PR. There are three review levels, Quick, Standard (the default) and Deep, and more thorough means more cost. It is not available to organizations with Zero Data Retention enabled or with the HIPAA configuration applied, and a review works only by reading the diff and the surrounding code. It does not build or run your code. My rule of thumb: if you are on Team or Enterprise and the only thing you want is reviews, use the managed product and delete the YAML you would otherwise maintain. If you want Claude to also take tickets, fix CI or run scheduled jobs, you need the action anyway, and you can run both. For a wider look at the review tool market, see the [AI code review tools comparison](/blog/ai-code-review-tools-comparison-2026/). Teach It Your Standards With CLAUDE.md Claude reads your CLAUDE.md on every run. That makes it the cheapest lever you have over review quality. Put in the things a human reviewer on your team would say out loud: code style rules, what counts as a blocker, patterns to prefer, directories nobody should touch. Two cautions from the docs and from common sense: - Keep it short. The docs list a concise CLAUDE.md as a cost saver precisely because it is read on every run. A 600-line file is a token tax on every pull request. - Move workflow detail into skills. Rules about how to do a task belong in a skill that loads on demand. Rules about what the project is belong in CLAUDE.md. For the managed Code Review product, REVIEW.md is read from the base or default branch and never from the pull request's own changes. That detail matters: it means a contributor cannot edit the review instructions inside the PR they want reviewed. Run Claude on a Schedule With a prompt and a schedule trigger, the action becomes a cron job that can think. The docs example generates a daily report of commits and open issues: name: Daily Report on: schedule: - cron: "0 9 * * *" jobs: report: runs-on: ubuntu-latest permissions: contents: read issues: read id-token: write steps: - uses: anthropics/claude-code-action@v1 with: anthropic_api_key: ${{ secrets.ANTHROPIC_API_KEY }} prompt: "Generate a summary of yesterday's commits and open issues" claude_args: | --model claude-opus-5-5 --allowedTools "mcp__github__list_commits,mcp__github__list_issues" Things to know before you rely on it: - No tools by default for a plain-text prompt. Claude has no shell or GitHub API access until you grant tools with --allowedTools or a permissions.allow rule in the settings input. A skill invocation can use the tools its own frontmatter grants. - Default branch only. GitHub runs scheduled workflows only from the default branch. - 60-day shutoff on public repos. GitHub disables the schedule after 60 days without repository activity. - Output goes to the run log unless you tell it to post somewhere. - The actor check still applies. GitHub attributes scheduled runs to a repository user, usually the one who last changed the cron line. If that user is a bot, list it in allowed_bots. Good scheduled jobs are boring ones: a weekly dependency report, a stale-issue triage summary, a changelog draft. Anything that mutates the repo on a timer deserves a second look. Who Can Trigger It, and What It Can Touch This is the part to read twice. You are letting text typed into a GitHub comment start a process that has write access to your code and a billable credential. The action has guardrails, and you should know exactly what they are. Two checks run before Claude starts. The run fails if either rejects the actor: - Write access. On issue and pull request events, the triggering user must have write access to the repo. To allow specific users without it, set allowed_non_write_users and pass your own github_token. - Human actor. On every event, a bot actor is rejected unless it is in allowed_bots. This is what stops bots from triggering Claude in a loop. The app has a wide permission set. This surprised me when I first read it. The Claude GitHub App is shared by every Claude feature that integrates with GitHub, so installing it grants Actions, Checks, Contents, Discussions, Issues, Pull requests, Workflows and Repository hooks at read and write, plus Administration, Members, Merge queues, Metadata and Statuses at read. GitHub does not let you accept a subset. If your organization only allows what the action itself needs (Contents, Issues, Pull requests), the docs say to create a custom GitHub App with just those. Note that a custom app covers only the action. Code Review and web auto-fix still need the official app. What I would do on top of the defaults: - Give every job the minimum permissions block. The review workflow above is read-only on purpose. Only the @claude workflow needs write. - Treat issue and PR text as untrusted input. Anyone who can comment can put instructions in front of the model. The write-access check limits who that is, but a repo with many collaborators still has many people who can steer it. This is my reasoning, not a quote from the docs: assume a hostile sentence will eventually reach the prompt, and make sure the damage is bounded by the token's permissions and --allowedTools. The [least-privilege guide for coding agents](/blog/coding-agent-permissions-least-privilege-2026/) goes deeper on that thinking. - Never put a key in the repo. Secrets only, referenced as ${{ secrets.ANTHROPIC_API_KEY }}. The docs repeat this one for a reason. - Review everything before merging. Branch protection should still require a human approval. Claude's PRs are PRs. - Pin the action. The docs use @v1. For a job that holds write permissions, pinning a full commit SHA is the standard GitHub hardening advice, and the cost is updating it on purpose. If you want a hard stop on specific commands inside the job, [Claude Code hooks](/blog/claude-code-hooks-guide-2026/) work in CI the same way they work on your machine. What a Run Costs and How to Cap It Each run spends two different things: - GitHub Actions minutes. The job runs on GitHub-hosted runners and uses your Actions minutes. - Tokens. Based on prompt and response length, task complexity and codebase size. With an API key you pay API rates. With an OAuth token, runs use your Claude subscription. The docs list the levers. I would turn most of them on from day one: - --max-turns in claude_args caps iterations. claude_args: "--max-turns 5 --model claude-sonnet-5" is the docs' own example. - Workflow timeout-minutes stops a runaway job. Pick a number you would be annoyed to exceed, say 15. - Concurrency controls limit parallel runs so five comments in a row do not start five jobs. A group keyed on the issue or PR number works: concurrency: group: claude-${{ github.event.issue.number || github.event.pull_request.number }} cancel-in-progress: false - A concise CLAUDE.md, since it is read on every run. - Specific requests and issue templates so Claude needs fewer turns to find the problem. - A cheaper model for routine jobs. Review and triage rarely need the largest model. Set it with --model in claude_args, or leave it off to use the Claude Code default. For org-wide tracking, the docs point to the analytics dashboard and the monitoring page. If you are on a subscription token and your allowance keeps vanishing, the reason is probably in [what drains Claude Code usage fastest](/blog/claude-code-usage-limits-2026/): long contexts, subagents and large models, all of which a busy CI job can hit. Roll It Out Across an Organization Quick setup and manual setup both work one repo at a time. For many repos: 1. Install the Claude GitHub App once at the organization level, on all repos or a selected list. 2. Store the secret as an organization-level Actions secret. Use an API key, not an OAuth token, for the reason covered earlier. 3. Add the workflow to each repo that should run it, or define the job once as a reusable workflow that each repo calls. The reusable route means one place to fix the version pin or the --max-turns value. Bedrock, Vertex and Foundry By default the action calls the Claude API directly. To route inference through your own cloud account instead, set one input: - use_bedrock: "true" for Amazon Bedrock - use_vertex: "true" for Google Cloud's Agent Platform - use_foundry: "true" for Microsoft Foundry All three authenticate through OIDC identity federation, so no static cloud credentials sit in the repository. The setup has its own page in the docs, including a custom GitHub App with APP_ID and APP_PRIVATE_KEY secrets, and I would follow that page step by step instead of improvising. Troubleshooting Claude does not respond to @claude. - Check the GitHub App is installed on the repo and workflows are enabled. - Check the secret exists in repository secrets. - Check the comment has @claude as a whole word. - Check the commenting user has write access. CI does not run on Claude's commits. GitHub does not trigger workflows on commits made with the default GITHUB_TOKEN. If you pass github_token: ${{ secrets.GITHUB_TOKEN }} to the action, remove it so the action authenticates as the Claude GitHub App, or pass a custom app token. Also check that your CI triggers include the events Claude's pushes produce, such as push or pull_request. This one bites almost everyone: Claude opens a PR, the checks never start, and the PR looks stuck. Authentication errors. Test the key or token locally with claude before debugging the workflow. A dead credential looks the same as a workflow bug from the outside. A review job passes but nothing shows on the PR. You are in automation mode without --comment and the inline comment tool. See the review section above. Still on @beta. Update to v1: change @beta to @v1, remove the mode input, replace direct_prompt with prompt, and move CLI options such as max_turns and model into claude_args. custom_instructions becomes --append-system-prompt. Where This Fits in Your Workflow My view is that the action earns its keep in two places and wastes money in a third. It earns its keep on small, well-specified tickets, where the fix is a paragraph of description and a test. Typos in docs, a missing null check, a dependency bump with a changelog to read. These are the jobs where a twenty-minute wait for a draft PR beats context-switching. It also earns its keep as a second reader on pull requests, especially on a team where review is the bottleneck. Used that way it is a filter that catches the obvious mistakes before a human spends attention. If review volume is your actual problem, the [AI code review bottleneck](/blog/ai-code-review-bottleneck-2026/) is worth a read before you add more generated code to the pile. It wastes money when the task is vague. "Improve the performance of the app" will burn turns, produce a sprawling diff and teach your team to distrust the bot. The fix is the same as in any Claude Code session: smaller tasks, clearer instructions. If you are new to the tool itself, start with [how to use Claude Code](/blog/how-to-use-claude-code-2026/) locally before you hand it your CI. A sensible rollout looks like this: 1. Run /install-github-app on one low-risk repo. 2. Start with the review workflow, read-only permissions, --max-turns set. 3. Add the @claude workflow once you trust the reviews. 4. Write the project rules into a short CLAUDE.md. 5. Add concurrency and a timeout. 6. Move to an org-level API key and a reusable workflow when a second repo wants it. Frequently Asked Questions What is Claude Code GitHub Actions? It is a GitHub Action, anthropics/claude-code-action, that runs Claude Code in your workflows. Mention @claude in a PR or issue comment and it analyzes code, implements changes and pushes commits. With a prompt it can run on any GitHub event. How do I install Claude Code on GitHub? Run /install-github-app inside Claude Code in your repository. It installs the Claude GitHub App, saves an authentication secret and opens a pull request with the workflow files. You need admin access and the GitHub CLI authenticated with gh auth login. Is Claude Code GitHub Actions free? The action is open source, but each run uses GitHub Actions minutes and either API tokens or your Claude subscription allowance. Cost depends on prompt length, task complexity and codebase size. Does Claude Code GitHub integration work with a Claude Pro or Max subscription? Yes. Generate a token with claude setup-token and store it as CLAUDE_CODE_OAUTH_TOKEN. Pro, Max, Team and Enterprise plans can do this. Runs then use your subscription instead of API billing. Should I use an API key or an OAuth token? Use an API key for shared and team repos. An OAuth token is tied to the subscription of the person who generated it, so it breaks when that person leaves and spends their personal allowance. How do I trigger Claude in a pull request? Comment @claude followed by your request, from a user with write access to the repository. The trigger phrase is configurable with trigger_phrase. Can Claude Code review pull requests automatically? Yes, two ways. Add a workflow that runs a review skill on pull_request events, with --comment so it posts to the PR. Or, on Team and Enterprise plans, turn on the managed Code Review product, which needs no workflow file. What is the difference between Claude Code Review and the GitHub Action? Code Review is a managed service on Anthropic infrastructure for Team and Enterprise plans. It reviews only, with read-only access. The action runs on your runners, works on any plan with a key, and can also edit code, push commits and run any prompt. Does Claude Code GitHub work on GitLab? /install-github-app works only with github.com repositories. For GitLab pipelines, use Claude Code GitLab CI/CD. Why is Claude not responding to my @claude comment? Check that the app is installed, workflows are enabled, the secret is set, the comment contains @claude as a full word, and the commenter has write access. Why does CI not run on Claude's commits? GitHub does not trigger workflows for commits made with the default GITHUB_TOKEN. Remove github_token: ${{ secrets.GITHUB_TOKEN }} from the action so it authenticates as the Claude GitHub App. Can Claude Code run on a schedule in GitHub Actions? Yes. Add a schedule trigger and a prompt. Scheduled workflows run only from the default branch, and public repos lose the schedule after 60 days without activity. How do I limit what Claude can do in CI? Set a minimal permissions block on the job, restrict tools with --allowedTools, cap iterations with --max-turns, add a workflow timeout, and require human approval before merge. Can I use Claude Code GitHub Actions with Bedrock or Vertex? Yes. Set use_bedrock, use_vertex or use_foundry to "true" and authenticate through OIDC federation instead of an API key. Does the managed Code Review work with Zero Data Retention? No. It is not available to organizations with Zero Data Retention enabled or the HIPAA configuration applied. Sources, all checked October 29, 2026: Claude Code GitHub Actions, Code Review, the claude-code-action repository with its setup guide and security notes, and the Claude GitHub App.

10 dépôts GitHub gratuits qui explosent en ce moment 🔗 LES 10 PROJETS DE LA VIDÉO 1. REA (Reverse Engineer Anything) ➜ morluto/rea 2. Les plugins officiels d'Anthropic ➜ anthropics/claude-plugins-official 3. HyperFrames ➜ heygen-com/hyperframes 4. VoiceStudio ➜ debpalash/VoiceStudio 5. ComfyUI ➜ Comfy-Org/ComfyUI (serveur MCP officiel : Comfy-Org/comfy-mcp) 6. gstack ➜ garrytan/gstack 7. OpenCode ➜ anomalyco/opencode 8. Orca ➜ stablyai/orca 9. Ruflo ➜ ruvnet/ruflo 10. L'audit de sécurité de Cloudflare ➜ cloudflare/security-audit-skill ⚠️ REA : installez-le uniquement depuis le dépôt officiel morluto/rea (ou rea.tools), des copies circulent sous des noms proches. REA, HyperFrames, gstack, ComfyUI… Voici 10 projets gratuits sur GitHub qui changent pas mal ce qu'on peut faire avec Claude et Claude Code en ce moment. Pour chacun, je vous montre à quoi il sert, comment il s'installe et où sont ses limites. ⏱️ CHAPITRES 0:00 Le repo dont tout le monde parle 0:27 1. REA 2:04 2. Les plugins officiels d'Anthropic 3:16 3. HyperFrames 4:39 4. VoiceStudio 5:25 5. ComfyUI 6:36 6. gstack 8:05 7. OpenCode 8:58 8. Orca 9:59 9. Ruflo 10:57 10. L'audit de sécurité de Cloudflare 📺 À VOIR AUSSI • Comment j'ai automatisé mon montage vidéo avec Claude Code ➜ youtube.com • 10 outils INDISPENSABLES pour créer un SaaS avec Claude Code ➜ youtube.com ✉️ Newsletter ➜ digitiz.fr #ClaudeCode #Claude #GitHub [en-US] 10 Free GitHub Repos Exploding Right Now 🔗 THE 10 PROJECTS FEATURED IN THE VIDEO 1. REA (Reverse Engineer Anything) ➜ morluto/rea 2. Anthropic's Official Plugins ➜ anthropics/claude-plugins-official 3. HyperFrames ➜ heygen-com/hyperframes 4. VoiceStudio ➜ debpalash/VoiceStudio 5. ComfyUI ➜ Comfy-Org/ComfyUI (Official MCP server: Comfy-Org/comfy-mcp) 6. gstack ➜ garrytan/gstack 7. OpenCode ➜ anomalyco/opencode 8. Orca ➜ stablyai/orca 9. Ruflo ➜ ruvnet/ruflo 10. Cloudflare Security Audit Skill ➜ cloudflare/security-audit-skill ⚠️ REA: Install it only from the official repository morluto/rea (or rea.tools). Copies are circulating under similar names. REA, HyperFrames, gstack, ComfyUI… Here are 10 free projects on GitHub that significantly change what you can do with Claude and Claude Code right now. For each one, I'll show you what it does, how to install it, and its limitations. ⏱️ CHAPTERS 0:00 The Repo Everyone's Talking About 0:27 1. REA 2:04 2. Anthropic's Official Plugins 3:16 3. HyperFrames 4:39 4. VoiceStudio 5:25 5. ComfyUI 6:36 6. gstack 8:05 7. OpenCode 8:58 8. Orca 9:59 9. Ruflo 10:57 10. Cloudflare's Security Audit 📺 ALSO WATCH • How I automated my video editing with Claude Code ➜ youtube.com • 10 ESSENTIAL tools for creating a SaaS with Claude Code ➜ youtube.com ✉️ Newsletter ➜ digitiz.fr #ClaudeCode #Claude #GitHub

Copy viral videos, change your TV’s OS, free MailChimp and all the top repos Link to Resource: thenextnewthing.ai Presented by Zapier: zapier.com Andrew Warner and Adam break down trending GitHub repositories and open-source AI tools of the week. From AI coding agents and security plugins to local voice cloning and shared memory systems, discover the projects developers are watching. They explore REA, a tool for recreating software features, and debate when inspiration becomes code theft. They also cover Claude's official security plugins, Ruflo's multi-agent coding teams, Android's official AI skills, scientific research skills, and OpenCode's open-source coding environment. Plus, see how CLI-Anything gives AI agents access to desktop apps, Cursor plugins extend coding workflows, and AI Memory helps agents share context across models and devices. The episode also features Mailcheap, an open-source newsletter alternative; CleanYourTV for customizing smart TVs; and Hypeit for recreating viral video formats. In the community showcase, Andrew and Adam highlight viewer-built projects, explore smarter AI model routing, and explain how secure containers help protect your computer and sensitive files. Repo featured: 1. REA — Recreate Software Features: morluto/rea 2. Claude Plugins Official — Security and Coding Skills: anthropics/claude-plugins-official 3. Ruflo — Multi-Agent Coding Teams: ruvnet/ruflo 4. Android Skills — Build Better Android Apps With AI: android/skills 5. Scientific Agent Skills — AI Skills for Research: K-Dense-AI/scientific-agent-skills 6. OpenCode — Open-Source AI Coding Agent: anomalyco/opencode 7. CLI-Anything — Connect AI Agents to Desktop Applications: HKUDS/CLI-Anything 8. Cursor Plugins — Skills, Tools, and Integrations: cursor/plugins 9. AI Memory — Shared Memory Across AI Agents: akitaonrails/ai-memory 10. VoiceStudio — Local AI Voice Cloning: debpalash/VoiceStudio Other Repos Discussed: - Mailcheap — An open-source, agent-friendly newsletter platform: marclou/mailcheap - CleanYourTV — Customize your smart TV and reduce unwanted bloatware: Gathin23/cleanyourtv - Hypeit — Recreate viral video formats with AI: hypit-ai/hypit First Star Repos: - Presenter — Generate and narrate presentations with AI: bisque-cloud/presenter - Delegate Router — Route tasks to the right AI model: pvjbuilds/delegate-router - This Thing Show — AI-powered book content ratings: R8rly/show - Kapathi Skills Extended — A coding agent that asks instead of guessing: anshuxinha/karpathy-skills-extended - Container — Run AI agents inside a Linux sandbox: aerovato/container 00:00 - This Week's Hottest GitHub Repositories 00:18 - Mailcheap: An Open-Source Alternative to Beehiiv 02:26 - CleanYourTV: Remove Smart TV Bloatware 03:41 - Hypeit: Recreate Viral Video Formats 05:29 - REA: Copying Software Features With AI 09:19 - Claude Plugins Official: Scan Code for Security Issues 11:21 - Ruflo: Build a Team of AI Coding Agents 12:54 - Zapier: Automate Workflows and Reduce AI Costs 14:25 - Android Skills: Help AI Build Better Apps 16:48 - Scientific Agent Skills for Researchers 18:12 - OpenCode: An Open-Source AI Coding Agent 20:37 - CLI-Anything: Give AI Agents Access to Desktop Apps 22:34 - The Road to 100,000 Subscribers 23:32 - Cursor Plugins and P-Stack Skills 24:40 - The Ketchup Skill: Catch Up on Your AI Conversations 25:01 - AI Memory: Share Context Across Agents and Devices 26:54 - VoiceStudio: Run Voice Cloning Locally 28:26 - Community Projects From Viewers 28:57 - Delegate Router: Choose the Right AI Model for Each Task 30:36 - AI-Powered Book Content Ratings 31:53 - Kapathi Skills Extended: An Agent That Asks Questions 32:11 - Container: Keep AI Agents in a Safe Sandbox 33:22 - Final Thoughts and Community Feedback 👉 Media/Sponsorship Inquiries: thenextnewthing.ai 👉 Join us: thenextnewthing.ai 🔥 Say hi to Adam: linkedin.com [de-DE] Virale Videos kopieren, das Betriebssystem deines Fernsehers ändern, kostenloses MailChimp und di... Link zur Ressource: thenextnewthing.ai Präsentiert von Zapier: zapier.com Andrew Warner und Adam analysieren die angesagtesten GitHub-Repositories und Open-Source-KI-Tools der Woche. Von KI-Codierungsagenten und Sicherheits-Plugins bis hin zu lokalem Voice-Cloning und Shared-Memory-Systemen – entdecken Sie die Projekte, die Entwickler im Blick haben. Sie erkunden REA, ein Tool zum Nachbilden von Softwarefunktionen, und diskutieren, wann Inspiration zu Code-Diebstahl wird. Außerdem behandeln sie Claudes offizielle Sicherheits-Plugins, Ruflos Multi-Agenten-Codierungsteams, Androids offizielle KI-Skills, wissenschaftliche Forschungs-Skills und OpenCodes Open-Source-Codierungsumgebung. Zusätzlich erfahren Sie, wie CLI-Anything KI-Agenten Zugriff auf Desktop-Anwendungen ermöglicht, Cursor-Plugins Codierungs-Workflows erweitern und AI Memory Agenten hilft, Kontext über Modelle und Geräte hinweg zu teilen. Die Folge stellt außerdem Mailcheap vor, eine Open-Source-Alternative für Newsletter, und CleanYourTV zur Personalisierung von Smart-TVs. und Hypeit für die Nachbildung viraler Videoformate. In der Community-Präsentation stellen Andrew und Adam von Zuschauern erstellte Projekte vor, zeigen intelligentere KI-Modell-Weiterleitung auf und erklären, wie sichere Container Ihren Computer und Ihre sensiblen Dateien schützen. Empfohlene Repositories: 1. REA – Softwarefunktionen nachbilden: morluto/rea 2. Claude Plugins Official – Sicherheits- und Programmierkenntnisse: anthropics/claude-plugins-official 3. Ruflo – Multi-Agenten-Programmierteams: ruvnet/ruflo 4. Android Skills – Bessere Android-Apps mit KI entwickeln: android/skills 5. Scientific Agent Skills – KI-Kenntnisse für die Forschung: K-Dense-AI/scientific-agent-skills 6. OpenCode – Open-Source-KI-Programmieragent: anomalyco/opencode 7. CLI-Anything – KI-Agenten mit Desktop-Anwendungen verbinden: HKUDS/CLI-Anything 8. Cursor Plugins – Fähigkeiten, Tools und Integrationen: cursor/plugins 9. AI Memory – Gemeinsamer Speicher für KI-Agenten: akitaonrails/ai-memory 10. VoiceStudio – Lokales KI-Stimmenklonen: debpalash/VoiceStudio Weitere erwähnte Repositories: - Mailcheap – Eine Open-Source-Newsletter-Plattform für Agenten: marclou/mailcheap - CleanYourTV – Smart-TV personalisieren und unerwünschte Bloatware entfernen: Gathin23/cleanyourtv - Hypeit – Virale Videoformate mit KI nachbilden: hypit-ai/hypit Erstklassige Repositories: - Presenter – Präsentationen mit KI erstellen und vertonen: bisque-cloud/presenter - Delegate Router – Aufgaben an das richtige KI-Modell weiterleiten: pvjbuilds/delegate-router - This Thing Show – KI-gestützte Buchbewertungen: R8rly/show - Kapathi Skills Extended – Ein Programmieragent, der fragt statt zu raten: anshuxinha/karpathy-skills-extended - Container – KI-Agenten in einer Linux-Sandbox ausführen: aerovato/container 00:00 – Die heißesten GitHub-Repositories dieser Woche 00:18 – Mailcheap: Eine Open-Source-Alternative zu Beehiiv 02:26 – CleanYourTV: Smart-TV-Bloatware entfernen 03:41 – Hypeit: Virale Videoformate nachbilden 05:29 – REA: Softwarefunktionen mit KI kopieren 09:19 – Claude Plugins Official: Code auf Sicherheitslücken prüfen 11:21 – Ruflo: Erstellen eines Team von KI-Programmieragenten 12:54 – Zapier: Workflows automatisieren und KI-Kosten senken 14:25 – Android Skills: KI beim Erstellen besserer Apps unterstützen 16:48 – Wissenschaftliche Agenten-Skills für Forscher 18:12 – OpenCode: Ein Open-Source-KI-Programmieragent 20:37 – CLI-Anything: KI-Agenten Zugriff auf Desktop-Anwendungen ermöglichen 22:34 – Der Weg zu 100.000 Abonnenten 23:32 – Cursor-Plugins und P-Stack-Skills 24:40 – Der Ketchup-Skill: KI-Konversationen im Blick behalten 25:01 – KI-Speicher: Kontext zwischen Agenten und Geräten teilen 26:54 – VoiceStudio: Sprachklonen lokal ausführen 28:26 – Community-Projekte von Zuschauern 28:57 – Delegate Router: Das richtige KI-Modell für jede Aufgabe auswählen 30:36 – KI-gestützte Buchinhalte Bewertungen 31:53 – Erweiterte Kapathi-Fähigkeiten: Ein Agent, der Fragen stellt 32:11 – Container: KI-Agenten in einer sicheren Umgebung halten 33:22 – Schlussbetrachtung und Feedback der Community 👉 Medien-/Sponsoringanfragen: thenextnewthing.ai 👉 Werden Sie Teil unseres Teams: thenextnewthing.ai 🔥 Kontaktieren Sie Adam: linkedin.com [es-US] Copia videos virales, cambia el SO de tu TV, MailChimp gratis y todos los mejores repositorios Enlace al recurso: thenextnewthing.ai Presentado por Zapier: zapier.com Andrew Warner y Adam analizan los repositorios de GitHub y las herramientas de IA de código abierto más populares de la semana. Desde agentes de codificación de IA y complementos de seguridad hasta clonación de voz local y sistemas de memoria compartida, descubre los proyectos que los desarrolladores siguen de cerca. Exploran REA, una herramienta para recrear funcionalidades de software, y debaten cuándo la inspiración se convierte en plagio de código. También abordan los complementos de seguridad oficiales de Claude, los equipos de codificación multiagente de Ruflo, las habilidades de IA oficiales de Android, las habilidades de investigación científica y el entorno de codificación de código abierto de OpenCode. Además, descubre cómo CLI-Anything permite a los agentes de IA acceder a aplicaciones de escritorio, cómo los complementos de Cursor amplían los flujos de trabajo de codificación y cómo AI Memory ayuda a los agentes a compartir contexto entre modelos y dispositivos. El episodio también incluye Mailcheap, una alternativa de boletín informativo de código abierto; CleanYourTV para personalizar televisores inteligentes; y Hypeit para recrear formatos de vídeo virales. En la muestra de la comunidad, Andrew y Adam destacan proyectos creados por los espectadores, exploran el enrutamiento inteligente de modelos de IA y explican cómo los contenedores seguros ayudan a proteger tu ordenador y tus archivos confidenciales. Repositorios destacados: 1. REA — Recrea funcionalidades de software: morluto/rea 2. Claude Plugins Official — Seguridad y habilidades de programación: anthropics/claude-plugins-official 3. Ruflo — Equipos de programación multiagente: ruvnet/ruflo 4. Android Skills — Crea mejores aplicaciones Android con IA: android/skills 5. Scientific Agent Skills — Habilidades de IA para la investigación: K-Dense-AI/scientific-agent-skills 6. OpenCode — Agente de programación de IA de código abierto: anomalyco/opencode 7. CLI-Anything — Conecta agentes de IA a aplicaciones de escritorio: HKUDS/CLI-Anything 8. Cursor Plugins — Habilidades, herramientas e integraciones: cursor/plugins 9. AI Memory — Memoria compartida entre agentes de IA: akitaonrails/ai-memory 10. VoiceStudio — Clonación de voz local con IA: debpalash/VoiceStudio Otros repositorios mencionados: - Mailcheap — Plataforma de boletines informativos de código abierto y fácil de usar para agentes: marclou/mailcheap - CleanYourTV — Personaliza tu Smart TV y reduce el software innecesario: Gathin23/cleanyourtv - Hypeit — Recrea formatos de vídeo virales con IA: hypit-ai/hypit Repositorios destacados: - Presenter — Genera y narra presentaciones con IA: bisque-cloud/presenter - Delegate Router — Enruta tareas al modelo de IA adecuado: pvjbuilds/delegate-router - This Thing Show: Calificaciones de contenido de libros con IA: R8rly/show - Kapathi Skills Extended: Un agente de programación que pregunta en lugar de adivinar: anshuxinha/karpathy-skills-extended - Container: Ejecuta agentes de IA dentro de un entorno aislado de Linux: aerovato/container 00:00 - Los repositorios de GitHub más populares de esta semana 00:18 - Mailcheap: Una alternativa de código abierto a Beehiiv 02:26 - CleanYourTV: Elimina el software innecesario de tu Smart TV 03:41 - Hypeit: Recrea formatos de vídeo virales 05:29 - REA: Copia de funciones de software con IA 09:19 - Claude Plugins Official: Analiza el código en busca de problemas de seguridad 11:21 - Ruflo: Compilación Un equipo de agentes de codificación de IA 12:54 - Zapier: Automatiza flujos de trabajo y reduce costes de IA 14:25 - Habilidades de Android: Ayuda a la IA a crear mejores apps 16:48 - Habilidades de agentes científicos para investigadores 18:12 - OpenCode: Un agente de codificación de IA de código abierto 20:37 - CLI-Anything: Proporciona a los agentes de IA acceso a aplicaciones de escritorio 22:34 - Camino a los 100 000 suscriptores 23:32 - Plugins de cursor y habilidades de P-Stack 24:40 - La habilidad Ketchup: Ponte al día con tus conversaciones con IA 25:01 - Memoria de IA: Comparte contexto entre agentes y dispositivos 26:54 - VoiceStudio: Ejecuta la clonación de voz localmente 28:26 - Proyectos de la comunidad de los espectadores 28:57 - Enrutador delegado: Elige el modelo de IA adecuado para cada tarea 30:36 - Contenido de libros con IA Calificaciones 31:53 - Habilidades de Kapathi ampliadas: Un agente que hace preguntas 32:11 - Contenedor: Mantén a los agentes de IA en un entorno seguro 33:22 - Reflexiones finales y comentarios de la comunidad... [fr-FR] Copiez des vidéos virales, changez l’OS de votre TV, un MailChimp gratuit et tous les meilleurs r... Lien vers la ressource : thenextnewthing.ai Présenté par Zapier : zapier.com Andrew Warner et Adam analysent les dépôts GitHub les plus populaires et les outils d'IA open source de la semaine. Des agents de programmation IA aux plugins de sécurité, en passant par le clonage vocal local et les systèmes de mémoire partagée, découvrez les projets suivis de près par les développeurs. Ils explorent REA, un outil permettant de recréer des fonctionnalités logicielles, et débattent de la frontière entre inspiration et plagiat. Ils abordent également les plugins de sécurité officiels de Claude, les équipes de programmation multi-agents de Ruflo, les compétences d'IA officielles d'Android, les compétences de recherche scientifique et l'environnement de programmation open source d'OpenCode. De plus, découvrez comment CLI-Anything donne accès aux applications de bureau aux agents d'IA, comment les plugins Cursor étendent les flux de travail de programmation et comment AI Memory aide les agents à partager le contexte entre les modèles et les appareils. Cet épisode présente également Mailcheap, une alternative open source aux newsletters, et CleanYourTV pour personnaliser les téléviseurs connectés. et Hypeit pour la recréation de formats vidéo viraux. Dans la section dédiée à la communauté, Andrew et Adam mettent en avant des projets réalisés par les utilisateurs, explorent des modèles de routage IA plus intelligents et expliquent comment les conteneurs sécurisés contribuent à protéger votre ordinateur et vos fichiers sensibles. Dépôts mis en avant : 1. REA — Recréer des fonctionnalités logicielles : morluto/rea 2. Claude Plugins Official — Compétences en sécurité et en programmation : anthropics/claude-plugins-official 3. Ruflo — Équipes de programmation multi-agents : ruvnet/ruflo 4. Android Skills — Créer de meilleures applications Android grâce à l’IA : android/skills 5. Scientific Agent Skills — Compétences en IA pour la recherche : K-Dense-AI/scientific-agent-skills 6. OpenCode — Agent de programmation IA open source : anomalyco/opencode 7. CLI-Anything — Connecter des agents IA à des applications de bureau : HKUDS/CLI-Anything 8. Cursor Plugins — Compétences, outils et intégrations cursor/plugins 9. AI Memory — Mémoire partagée entre agents IA : akitaonrails/ai-memory 10. VoiceStudio — Clonage vocal IA local : debpalash/VoiceStudio Autres dépôts mentionnés : - Mailcheap — Plateforme de newsletter open source et compatible avec les agents : marclou/mailcheap - CleanYourTV — Personnalisez votre Smart TV et réduisez les logiciels inutiles : Gathin23/cleanyourtv - Hypeit — Recréez des formats vidéo viraux grâce à l’IA : hypit-ai/hypit Premiers dépôts recommandés : - Presenter — Générez et commentez des présentations avec l’IA : bisque-cloud/presenter - Delegate Router — Acheminez les tâches vers le modèle d’IA approprié : pvjbuilds/delegate-router - This Thing Show — Évaluation du contenu des livres par l'IA : R8rly/show - Kapathi Skills Extended — Un agent de programmation qui pose des questions au lieu de deviner : anshuxinha/karpathy-skills-extended - Container — Exécuter des agents d'IA dans un environnement isolé Linux : aerovato/container 00:00 - Les dépôts GitHub les plus populaires de la semaine 00:18 - Mailcheap : Une alternative open source à Beehiiv 02:26 - CleanYourTV : Supprimer les logiciels inutiles des téléviseurs connectés 03:41 - Hypeit : Recréer des formats vidéo viraux 05:29 - REA : Copier des fonctionnalités logicielles grâce à l'IA 09:19 - Claude Plugins Official : Analyser le code pour détecter les failles de sécurité 11:21 - Ruflo : Constituer une équipe Agents de codage IA 12:54 - Zapier : Automatisez vos flux de travail et réduisez les coûts de l’IA 14:25 - Compétences Android : Aidez l’IA à créer de meilleures applications 16:48 - Compétences d’agents scientifiques pour les chercheurs 18:12 - OpenCode : Un agent de codage IA open source 20:37 - CLI-Anything : Donnez aux agents IA accès aux applications de bureau 22:34 - Objectif : 100 000 abonnés 23:32 - Plugins de curseur et compétences P-Stack 24:40 - La compétence Ketchup : Suivez vos conversations avec l’IA 25:01 - Mémoire IA : Partagez le contexte entre agents et appareils 26:54 - VoiceStudio : Clonage vocal en local 28:26 - Projets de la communauté 28:57 - Delegate Router : Choisissez le modèle d’IA adapté à chaque tâche 30:36 - Contenu de livres enrichi par l’IA Évaluations 31:53 - Kapathi Skills Extended : Un agent qui pose des questions 32:11 - Container : Sécuriser les agents IA dans un environnement isolé 33:... [hi] वायरल वीडियो कॉपी करें, अपने टीवी का OS बदलें, मुफ्त Mailchimp और सभी टॉप रिपोस संसाधन का लिंक: thenextnewthing.ai प्रस्तुतकर्ता: Zapier: zapier.com एंड्रयू वार्नर और एडम इस सप्ताह के ट्रेंडिंग GitHub रिपॉजिटरी और ओपन-सोर्स AI टूल्स का विश्लेषण करते हैं। AI कोडिंग एजेंट और सुरक्षा प्लगइन्स से लेकर लोकल वॉयस क्लोनिंग और शेयर्ड मेमोरी सिस्टम तक, जानिए वे प्रोजेक्ट जिन पर डेवलपर्स की नजर है। वे सॉफ्टवेयर फीचर्स को रीक्रिएट करने वाले टूल REA पर चर्चा करते हैं और इस बात पर विचार-विमर्श करते हैं कि प्रेरणा कब कोड की चोरी बन जाती है। वे क्लाउड के आधिकारिक सुरक्षा प्लगइन्स, रूफ्लो की मल्टी-एजेंट कोडिंग टीमों, एंड्रॉइड के आधिकारिक AI स्किल्स, वैज्ञानिक अनुसंधान स्किल्स और ओपनकोड के ओपन-सोर्स कोडिंग एनवायरनमेंट पर भी बात करते हैं। इसके अलावा, देखिए कैसे CLI-Anything AI एजेंट्स को डेस्कटॉप ऐप्स तक पहुंच प्रदान करता है, कर्सर प्लगइन्स कोडिंग वर्कफ़्लो को बढ़ाते हैं, और AI मेमोरी एजेंट्स को मॉडल्स और डिवाइसेस में कॉन्टेक्स्ट साझा करने में मदद करती है। इस एपिसोड में ओपन-सोर्स न्यूज़लेटर विकल्प Mailcheap; स्मार्ट टीवी को कस्टमाइज़ करने के लिए CleanYourTV; और वायरल वीडियो फॉर्मेट को फिर से बनाने के लिए हाइपिट का धन्यवाद। कम्युनिटी शोकेस में, एंड्रयू और एडम दर्शकों द्वारा निर्मित परियोजनाओं पर प्रकाश डालते हैं, स्मार्ट एआई मॉडल रूटिंग की पड़ताल करते हैं, और बताते हैं कि सुरक्षित कंटेनर आपके कंप्यूटर और संवेदनशील फाइलों की सुरक्षा में कैसे मदद करते हैं। प्रमुख रिपॉजिटरी: 1. REA — सॉफ्टवेयर सुविधाओं को पुनः बनाएं: morluto/rea 2. Claude Plugins Official — सुरक्षा और कोडिंग कौशल: anthropics/claude-plugins-official 3. Ruflo — मल्टी-एजेंट कोडिंग टीमें: ruvnet/ruflo 4. Android Skills — AI के साथ बेहतर Android ऐप्स बनाएं: android/skills 5. Scientific Agent Skills — अनुसंधान के लिए AI कौशल: K-Dense-AI/scientific-agent-skills 6. OpenCode — ओपन-सोर्स AI कोडिंग एजेंट: anomalyco/opencode 7. CLI-Anything — AI एजेंटों को डेस्कटॉप एप्लिकेशन से कनेक्ट करें: HKUDS/CLI-Anything 8. Cursor Plugins — कौशल, उपकरण और एकीकरण: cursor/plugins 9. AI मेमोरी — AI एजेंटों के बीच साझा मेमोरी: akitaonrails/ai-memory 10. VoiceStudio — स्थानीय AI वॉइस क्लोनिंग: debpalash/VoiceStudio चर्चा किए गए अन्य रिपॉजिटरी: - Mailcheap — एक ओपन-सोर्स, एजेंट-अनुकूल न्यूज़लेटर प्लेटफ़ॉर्म: marclou/mailcheap - CleanYourTV — अपने स्मार्ट टीवी को कस्टमाइज़ करें और अनावश्यक ब्लोटवेयर को कम करें: Gathin23/cleanyourtv - Hypeit — AI के साथ वायरल वीडियो फ़ॉर्मेट को रीक्रिएट करें: hypit-ai/hypit फर्स्ट स्टार रिपॉजिटरी: - Presenter — AI के साथ प्रेजेंटेशन जेनरेट और नैरेट करें: bisque-cloud/presenter - Delegate Router — कार्यों को सही AI मॉडल तक पहुंचाएं: pvjbuilds/delegate-router - दिस थिंग शो — AI-आधारित पुस्तक सामग्री रेटिंग: R8rly/show - कपाथी स्किल्स एक्सटेंडेड — एक कोडिंग एजेंट जो अनुमान लगाने के बजाय प्रश्न पूछता है: anshuxinha/karpathy-skills-extended - कंटेनर — लिनक्स सैंडबॉक्स के अंदर AI एजेंट चलाएं: aerovato/container 00:00 - इस सप्ताह के सबसे चर्चित GitHub रिपॉजिटरी 00:18 - मेलचीप: बीहीव का एक ओपन-सोर्स विकल्प 02:26 - क्लीन योर टीवी: स्मार्ट टीवी ब्लोटवेयर हटाएं 03:41 - हाइपिट: वायरल वीडियो फॉर्मेट को रीक्रिएट करें 05:29 - आरईए: AI के साथ सॉफ्टवेयर फीचर्स कॉपी करना 09:19 - क्लाउड प्लगइन्स ऑफिशियल: सुरक्षा के लिए कोड स्कैन करें विषय 11:21 - रुफ्लो: एआई कोडिंग एजेंटों की टीम बनाएं 12:54 - जैपियर: वर्कफ़्लो को स्वचालित करें और एआई लागत कम करें 14:25 - एंड्रॉइड स्किल्स: एआई को बेहतर ऐप्स बनाने में मदद करें 16:48 - शोधकर्ताओं के लिए वैज्ञानिक एजेंट स्किल्स 18:12 - ओपनकोड: एक ओपन-सोर्स एआई कोडिंग एजेंट 20:37 - CLI-Anything: एआई एजेंटों को डेस्कटॉप ऐप्स तक पहुंच प्रदान करें 22:34 - 100,000 सब्सक्राइबर्स तक का सफर 23:32 - कर्सर प्लगइन्स और पी-स्टैक स्किल्स 24:40 - केचप स्किल: अपनी एआई बातचीत पर नज़र रखें 25:01 - एआई मेमोरी: एजेंटों और उपकरणों के बीच संदर्भ साझा करें 26:54 - वॉयसस्टूडियो: स्थानीय रूप से वॉयस क्लोनिंग चलाएं 28:26 - दर्शकों के सामुदायिक प्रोजेक्ट 28:57 - डेलीगेट राउटर: प्रत्येक के लिए सही एआई मॉडल चुनें कार्य 30:36 - एआई-आधारित पुस्तक सामग्री रेटिंग 31:53 - कपाथी स्किल्स एक्सटेंडेड: एक एजेंट जो प्रश्न पूछता है 32:11 - कंटेनर: एआई एजेंटों को सुरक्षित सैंडबॉक्स में रखें 33:22 - अंतिम विचार और सामुदायिक प्रतिक्रिया 👉 मीडिया/प्रायोजन संबंधी पूछताछ: thenextnewthing.ai 👉 हमसे जुड़ें: thenextnewthing.ai 🔥 एडम से संपर्क करें: linkedin.com [id] Salin video viral, ubah OS TV Anda, MailChimp gratis, dan semua repo terpopuler Tautan ke Sumber Daya: thenextnewthing.ai Dipersembahkan oleh Zapier: zapier.com Andrew Warner dan Adam membahas repositori GitHub dan alat AI sumber terbuka yang sedang tren minggu ini. Mulai dari agen pengkodean AI dan plugin keamanan hingga kloning suara lokal dan sistem memori bersama, temukan proyek-proyek yang dipantau oleh para pengembang. Mereka mengeksplorasi REA, alat untuk menciptakan kembali fitur perangkat lunak, dan memperdebatkan kapan inspirasi menjadi pencurian kode. Mereka juga membahas plugin keamanan resmi Claude, tim pengkodean multi-agen Ruflo, keterampilan AI resmi Android, keterampilan penelitian ilmiah, dan lingkungan pengkodean sumber terbuka OpenCode. Selain itu, lihat bagaimana CLI-Anything memberi agen AI akses ke aplikasi desktop, plugin Cursor memperluas alur kerja pengkodean, dan AI Memory membantu agen berbagi konteks di seluruh model dan perangkat. Episode ini juga menampilkan Mailcheap, alternatif buletin sumber terbuka; CleanYourTV untuk menyesuaikan TV pintar; dan Hypeit untuk menciptakan kembali format video viral. Dalam acara komunitas, Andrew dan Adam menyoroti proyek-proyek yang dibuat oleh pemirsa, mengeksplorasi perutean model AI yang lebih cerdas, dan menjelaskan bagaimana kontainer yang aman membantu melindungi komputer dan file sensitif Anda. Repositori unggulan: 1. REA — Recreate Software Features: morluto/rea 2. Claude Plugins Official — Security and Coding Skills: anthropics/claude-plugins-official 3. Ruflo — Multi-Agent Coding Teams: ruvnet/ruflo 4. Android Skills — Build Better Android Apps With AI: android/skills 5. Scientific Agent Skills — AI Skills for Research: K-Dense-AI/scientific-agent-skills 6. OpenCode — Open-Source AI Coding Agent: anomalyco/opencode 7. CLI-Anything — Connect AI Agents to Desktop Applications: HKUDS/CLI-Anything 8. Cursor Plugins — Skills, Tools, and Integrations: cursor/plugins 9. AI Memory — Memori Bersama Antar Agen AI: akitaonrails/ai-memory 10. VoiceStudio — Pengkloningan Suara AI Lokal: debpalash/VoiceStudio Repositori Lain yang Dibahas: - Mailcheap — Platform buletin sumber terbuka yang ramah agen: marclou/mailcheap - CleanYourTV — Sesuaikan TV pintar Anda dan kurangi bloatware yang tidak diinginkan: Gathin23/cleanyourtv - Hypeit — Buat ulang format video viral dengan AI: hypit-ai/hypit Repositori Bintang Pertama: - Presenter — Hasilkan dan narasikan presentasi dengan AI: bisque-cloud/presenter - Delegate Router — Arahkan tugas ke model AI yang tepat: pvjbuilds/delegate-router - This Thing Show — Peringkat konten buku bertenaga AI: R8rly/show - Kapathi Skills Extended — Agen pengkodean yang bertanya alih-alih menebak: anshuxinha/karpathy-skills-extended - Container — Jalankan agen AI di dalam sandbox Linux: aerovato/container 00:00 - Repositori GitHub Terpopuler Minggu Ini 00:18 - Mailcheap: Alternatif Sumber Terbuka untuk Beehiiv 02:26 - CleanYourTV: Hapus Bloatware Smart TV 03:41 - Hypeit: Buat Ulang Format Video Viral 05:29 - REA: Menyalin Fitur Perangkat Lunak dengan AI 09:19 - Claude Plugins Official: Pindai Kode untuk Masalah Keamanan 11:21 - Ruflo: Bangun Tim Pengkodean AI Agen 12:54 - Zapier: Otomatiskan Alur Kerja dan Kurangi Biaya AI 14:25 - Android Skills: Bantu AI Membangun Aplikasi yang Lebih Baik 16:48 - Keterampilan Agen Ilmiah untuk Peneliti 18:12 - OpenCode: Agen Pengkodean AI Sumber Terbuka 20:37 - CLI-Anything: Berikan Akses Agen AI ke Aplikasi Desktop 22:34 - Jalan Menuju 100.000 Pelanggan 23:32 - Plugin Kursor dan Keterampilan P-Stack 24:40 - Keterampilan Saus Tomat: Ikuti Percakapan AI Anda 25:01 - Memori AI: Bagikan Konteks di Seluruh Agen dan Perangkat 26:54 - VoiceStudio: Jalankan Pengkloningan Suara Secara Lokal 28:26 - Proyek Komunitas dari Pemirsa 28:57 - Delegate Router: Pilih Model AI yang Tepat untuk Setiap Tugas 30:36 - Peringkat Konten Buku Bertenaga AI 31:53 - Kemampuan Kapathi yang Diperluas: Agen yang Mengajukan Pertanyaan 32:11 - Kontainer: Menjaga Agen AI dalam Sandbox yang Aman 33:22 - Pemikiran Akhir dan Umpan Balik Komunitas 👉 Pertanyaan Media/Sponsor: thenextnewthing.ai 👉 Bergabunglah dengan kami: thenextnewthing.ai 🔥 Sapa Adam: linkedin.com [it] Copia video virali, cambia il sistema operativo della tua TV, MailChimp gratuito e tutti i miglio... Link alla risorsa: thenextnewthing.ai Presentato da Zapier: zapier.com Andrew Warner e Adam analizzano i repository GitHub di tendenza e gli strumenti di intelligenza artificiale open source della settimana. Dagli agenti di programmazione IA e plugin di sicurezza alla clonazione vocale locale e ai sistemi di memoria condivisa, scoprite i progetti che gli sviluppatori stanno seguendo. Esplorano REA, uno strumento per ricreare funzionalità software, e discutono di quando l'ispirazione si trasforma in furto di codice. Parlano anche dei plugin di sicurezza ufficiali di Claude, dei team di programmazione multi-agente di Ruflo, delle competenze IA ufficiali di Android, delle competenze di ricerca scientifica e dell'ambiente di programmazione open source di OpenCode. Inoltre, scoprite come CLI-Anything consente agli agenti IA di accedere alle applicazioni desktop, come i plugin di Cursor estendono i flussi di lavoro di programmazione e come AI Memory aiuta gli agenti a condividere il contesto tra modelli e dispositivi. L'episodio presenta anche Mailcheap, un'alternativa open source alle newsletter; CleanYourTV per personalizzare le smart TV; e Hypeit per ricreare formati video virali. Nella vetrina della community, Andrew e Adam mettono in evidenza i progetti creati dagli spettatori, esplorano modelli di routing basati sull'intelligenza artificiale più intelligenti e spiegano come i container sicuri aiutano a proteggere il computer e i file sensibili. Repository in evidenza: 1. REA — Ricrea funzionalità software: morluto/rea 2. Claude Plugins Official — Competenze di sicurezza e programmazione: anthropics/claude-plugins-official 3. Ruflo — Team di programmazione multi-agente: ruvnet/ruflo 4. Android Skills — Crea app Android migliori con l'IA: android/skills 5. Scientific Agent Skills — Competenze di IA per la ricerca: K-Dense-AI/scientific-agent-skills 6. OpenCode — Agente di programmazione IA open source: anomalyco/opencode 7. CLI-Anything — Collega agenti IA ad applicazioni desktop: HKUDS/CLI-Anything 8. Cursor Plugins — Competenze, strumenti e Integrazioni: cursor/plugins 9. AI Memory — Memoria condivisa tra agenti AI: akitaonrails/ai-memory 10. VoiceStudio — Clonazione vocale locale tramite AI: debpalash/VoiceStudio Altri repository discussi: - Mailcheap — Una piattaforma di newsletter open source e intuitiva per gli agenti: marclou/mailcheap - CleanYourTV — Personalizza la tua smart TV e riduci il bloatware indesiderato: Gathin23/cleanyourtv - Hypeit — Ricrea formati video virali con l'IA: hypit-ai/hypit Primi repository consigliati: - Presenter — Genera e narra presentazioni con l'IA: bisque-cloud/presenter - Delegate Router — Instrada le attività al modello AI corretto: Italiano: pvjbuilds/delegate-router - This Thing Show — Valutazione del contenuto dei libri basata sull'IA: R8rly/show - Kapathi Skills Extended — Un agente di programmazione che chiede invece di indovinare: anshuxinha/karpathy-skills-extended - Container — Esegui agenti IA all'interno di una sandbox Linux: aerovato/container 00:00 - I repository GitHub più popolari di questa settimana 00:18 - Mailcheap: un'alternativa open source a Beehiiv 02:26 - CleanYourTV: rimuove il bloatware dalle Smart TV 03:41 - Hypeit: ricrea i formati video virali 05:29 - REA: copia le funzionalità del software con l'IA 09:19 - Claude Plugins Official: scansiona il codice per problemi di sicurezza 11:21 - Ruflo: Crea un team di agenti di programmazione IA 12:54 - Zapier: Automatizza i flussi di lavoro e riduci i costi dell'IA 14:25 - Competenze Android: Aiuta l'IA a creare app migliori 16:48 - Competenze di agenti scientifici per i ricercatori 18:12 - OpenCode: Un agente di programmazione IA open source 20:37 - CLI-Anything: Dai agli agenti IA accesso alle app desktop 22:34 - La strada verso i 100.000 iscritti 23:32 - Plugin Cursor e competenze P-Stack 24:40 - La competenza Ketchup: Recupera le tue conversazioni IA 25:01 - Memoria IA: Condividi il contesto tra agenti e dispositivi 26:54 - VoiceStudio: Esegui la clonazione vocale in locale 28:26 - Progetti della community degli spettatori 28:57 - Delegate Router: Scegli il modello IA giusto per ogni attività 30:36 - Valutazione dei contenuti dei libri tramite IA 31:53 - Kapathi Skills Extended: un agente che pone domande 32:11 - Container: mantenere gli agenti IA in un ambiente protetto 33:22 - Considerazioni finali e feedback della community 👉 Richieste di sponsorizzazione/media: thenextnewthing.ai 👉 Unisciti a noi: thenextnewthing.ai 🔥 Saluta Adam: linkedin.com [pt-BR] Copie vídeos virais, mude o sistema da sua TV, MailChimp grátis e todos os melhores repositórios Link para o recurso: thenextnewthing.ai Apresentado por Zapier: zapier.com Andrew Warner e Adam analisam os repositórios do GitHub e as ferramentas de IA de código aberto em alta da semana. De agentes de codificação de IA e plugins de segurança a clonagem de voz local e sistemas de memória compartilhada, descubra os projetos que os desenvolvedores estão acompanhando. Eles exploram o REA, uma ferramenta para recriar recursos de software, e debatem quando a inspiração se torna plágio. Eles também abordam os plugins de segurança oficiais do Claude, as equipes de codificação multiagente do Ruflo, as habilidades oficiais de IA do Android, habilidades de pesquisa científica e o ambiente de codificação de código aberto do OpenCode. Além disso, veja como o CLI-Anything dá aos agentes de IA acesso a aplicativos de desktop, os plugins Cursor estendem os fluxos de trabalho de codificação e o AI Memory ajuda os agentes a compartilhar contexto entre modelos e dispositivos. O episódio também apresenta o Mailcheap, uma alternativa de código aberto para newsletters; o CleanYourTV para personalizar smart TVs; e o Hypeit para recriar formatos de vídeos virais. Na apresentação da comunidade, Andrew e Adam destacam projetos criados por espectadores, exploram roteamento de modelos de IA mais inteligentes e explicam como contêineres seguros ajudam a proteger seu computador e arquivos confidenciais. Repositórios em destaque: 1. REA — Recrie recursos de software: morluto/rea 2. Claude Plugins Official — Segurança e habilidades de programação: anthropics/claude-plugins-official 3. Ruflo — Equipes de programação multiagente: ruvnet/ruflo 4. Android Skills — Crie aplicativos Android melhores com IA: android/skills 5. Scientific Agent Skills — Habilidades de IA para pesquisa: K-Dense-AI/scientific-agent-skills 6. OpenCode — Agente de programação de IA de código aberto: anomalyco/opencode 7. CLI-Anything — Conecte agentes de IA a aplicativos de desktop: HKUDS/CLI-Anything 8. Cursor Plugins — Habilidades, ferramentas e integrações: cursor/plugins 9. AI Memory — Memória compartilhada entre agentes de IA: akitaonrails/ai-memory 10. VoiceStudio — Clonagem local de voz por IA: debpalash/VoiceStudio Outros repositórios mencionados: - Mailcheap — Uma plataforma de newsletter de código aberto e amigável para agentes: marclou/mailcheap - CleanYourTV — Personalize sua smart TV e reduza o bloatware indesejado: Gathin23/cleanyourtv - Hypeit — Recrie formatos de vídeos virais com IA: hypit-ai/hypit Primeiros repositórios em destaque: - Presenter — Gere e narre apresentações com IA: bisque-cloud/presenter - Delegate Router — Direcione tarefas para o modelo de IA correto: pvjbuilds/delegate-router - This Thing Show — Avaliações de conteúdo de livros com IA: R8rly/show - Kapathi Skills Extended — Um agente de codificação que pergunta em vez de adivinhar: anshuxinha/karpathy-skills-extended - Container — Execute agentes de IA dentro de um sandbox Linux: aerovato/container 00:00 - Os repositórios mais populares do GitHub desta semana 00:18 - Mailcheap: Uma alternativa de código aberto ao Beehiiv 02:26 - CleanYourTV: Remova o bloatware da sua Smart TV 03:41 - Hypeit: Recrie formatos de vídeos virais 05:29 - REA: Copiando recursos de software com IA 09:19 - Claude Plugins Official: Analise o código em busca de problemas de segurança 11:21 - Ruflo: Monte uma equipe de codificação com IA Agentes 12:54 - Zapier: Automatize fluxos de trabalho e reduza os custos de IA 14:25 - Habilidades do Android: Ajude a IA a criar aplicativos melhores 16:48 - Habilidades de agentes científicos para pesquisadores 18:12 - OpenCode: Um agente de codificação de IA de código aberto 20:37 - CLI-Anything: Dê acesso a aplicativos de desktop para agentes de IA 22:34 - O caminho para 100.000 inscritos 23:32 - Plugins de cursor e habilidades P-Stack 24:40 - A habilidade Ketchup: Acompanhe suas conversas com IA 25:01 - Memória de IA: Compartilhe contexto entre agentes e dispositivos 26:54 - VoiceStudio: Execute clonagem de voz localmente 28:26 - Projetos da comunidade de espectadores 28:57 - Delegate Router: Escolha o modelo de IA certo para cada tarefa 30:36 - Avaliações de conteúdo de livros com IA 31:53 - Kapathi Habilidades Ampliadas: Um Agente Que Faz Perguntas 32:11 - Contêiner: Mantenha os Agentes de IA em um Ambiente Seguro 33:22 - Considerações Finais e Feedback da Comunidade 👉 Consultas de Mídia/Patrocínio: thenextnewthing.ai 👉 Junte-se a nós: thenextnewthing.ai 🔥 Diga olá para Adam: linkedin.com

Anthropic's official plugin directory for Claude Code puts MCP servers, slash commands, agents and skills one command away. 🔗 anthropics/claude-plugins-official #claudepluginsofficial #claudecode #anthropic

이번 주 개발자라면 봐야 할 오픈소스 Top 10 +@ 총정리 | GitHub 트렌딩 (10월 3주) 이번 주(10/14~10/21) GitHub 트렌딩 주간 순위 Top 10과 순위 밖 주목 저장소 3개를, 레포마다 무엇을 하는지·이번 주 바뀐 것·쓸 때 주의할 점까지 정리했어요. 순위는 GitHub Trending(weekly) 페이지 순서 그대로예요(2026-10-21 06:00 KST 수집). 별 수는 수집 시점 값이에요. ▶ 챕터 0:00 오프닝 0:20 10위 android/skills 0:54 9위 ruflo 1:34 8위 VoiceStudio 2:11 7위 HyperFrames 2:46 6위 scientific-agent-skills 3:22 5위 Ghidra 3:55 4위 context-mode 4:33 3위 claude-plugins-official 5:15 2위 ai-job-search 6:01 번외 ① OpenCode 7:12 번외 ② vLLM 7:50 번외 ③ cursor/plugins 8:18 1위 REA 9:22 이번 주 확인할 것 ▶ Top 10 10위 android/skills — Google Android 팀 공식 에이전트 스킬 30개. 새 스킬 4개(LiteRT 온디바이스 AI·Compose 성능 등), Claude Code 플러그인 설치 9위 ruvnet/ruflo — Claude Code·Codex 위 다중 에이전트 도구(옛 claude-flow). v3.57.0: Codex 기본 권한 danger-full-access → workspace-write·on-request, 경로 우회(CWE-22) 수정, 벡터 검색 실제 연결 8위 debpalash/VoiceStudio — 로컬 음성 복제·더빙 앱(AGPL-3.0). 모바일 원격 앱 미리보기(소스 빌드), 데이터·모델 저장 위치 변경. 기본 음성 모델 상업 이용 조건은 별도 확인 7위 heygen-com/hyperframes — HTML로 쓰고 영상으로 렌더. 릴리스 27번: hyperframes audio(음량·무음 측정), 720p·480p 렌더, 비영어 받아쓰기 수정. ⚠ v0.8.164부터 웹사이트 캡처가 HTTPS_PROXY를 쓰지 않음(프록시만 쓰는 회사망은 확인) 6위 K-Dense-AI/scientific-agent-skills — 연구용 스킬 177개, 데이터베이스 80개 조회법, 스킬별 워크플로 도해. README: 한꺼번에 전부 설치하지 말 것 5위 NationalSecurityAgency/ghidra — NSA 연구국의 무료 리버스 엔지니어링 도구(Apache-2.0). Ghidra Server 인증서 검사 수정, 다음 버전 12.2는 JDK 25 필요 4위 mksglu/context-mode — 도구 출력을 따로 처리해 컨텍스트를 아끼는 플러그인. "315KB→5.4KB, 98%"는 개발사 발표. Elastic License 2.0(사내 사용 가능, 호스팅 서비스 재제공 불가) 3위 anthropics/claude-plugins-official — Claude Code 공식 플러그인 목록(315개). security-guidance: 커밋 리뷰어 읽기 전용, Claude Code 자격증명 파일 검토 제외. 외부 플러그인 내용은 Anthropic이 검증하지 않음(README) 2위 MadsLorentzen/ai-job-search — Claude Code 구직 도구. v1.7.3: 외부 업로드·비밀 파일 읽기 deny, git push ask, 공고 읽는 서브에이전트 읽기 전용, 숨은 지시문 필터. 포크는 항상 공개 → 개인용은 private 저장소 1위 morluto/rea — 코딩 에이전트를 Ghidra·Hopper·IDA에 연결하는 리버스 엔지니어링 도구(MIT). 6.5.0~6.7.0: 패킷·SQLite·WASM 분석, BREAKING 7건(버전 고정 권장). 공식 출처(morluto/rea, npm rea-agents)만 사용. ⚠ 타사 소프트웨어 분석은 약관(EULA)·저작권 문제. Claude Code·Codex·Cursor가 이런 작업을 막는다는 사용자 보고(REA Discussion #2095·#2027) ▶ 번외(순위 밖) anomalyco/opencode — 모델 회사에 묶이지 않는 오픈소스 코딩 에이전트. OpenCode 자체가 하네스(도구·반복은 내 컴퓨터, 모델 API로는 추론만). 2.0 발표, 기간 한정 무료 모델(DeepSeek V4.1 Flash Free 등, Zen 게이트웨이 경유). ⚠ 10/17 Console 보안 사고 → 외부 API 키 교체, 되도록 격리된 환경에서 사용. 기본 설정은 도구 승인 없음 vllm-project/vllm — 공개 언어 모델을 OpenAI 호환 API로 띄우는 엔진. Ollama와 비슷하지만 서빙 처리량(연속 배칭)에 강해 실제 서비스용. v0.32.0: 기본 CORS 해제(--allowed-origins), Python 3.11 이상 cursor/plugins(주간 12위) — 플러그인 101 → 112개(업무 SaaS 연동, 다수 쓰기 권한), pstack을 Claude Code에서도 설치 ▶ 끄는 방법·확인할 설정 - hyperframes: HYPERFRAMES_NO_TELEMETRY=1 - vLLM: VLLM_NO_USAGE_STATS=1 또는 DO_NOT_TRACK=1 - ruflo 3.57.0 이전 init: .codex/config.toml의 danger-full-access 확인 - security-guidance LLM 리뷰 끄기: ENABLE_CODE_SECURITY_REVIEW=0 - Android CLI 사용 통계: 명령마다 --no-metrics ▶ 관련 영상 - 이번 주 트렌딩 쇼츠(Top 5 요약): youtu.be - REA 직접 써 본 쇼츠: youtu.be - REA 실험 롱폼: youtu.be ▶ 링크 morluto/rea MadsLorentzen/ai-job-search anthropics/claude-plugins-official mksglu/context-mode NationalSecurityAgency/ghidra K-Dense-AI/scientific-agent-skills heygen-com/hyperframes debpalash/VoiceStudio ruvnet/ruflo android/skills anomalyco/opencode vllm-project/vllm cursor/plugins ※ 내레이션은 AI 보이스(ElevenLabs)예요. 실화면이 없는 부분은 공식 문서에 적힌 동작을 그린 원리 모션이에요. #GitHub트렌딩 #오픈소스 #개발자 #코딩에이전트 #ClaudeCode #SoloStack [en] This Week's Must-See Open Source: GitHub Trending Top 10 + Extras (Oct, Week 3) A full walkthrough of this week's GitHub Trending (weekly) Top 10 in page order (collected 2026-10-21 06:00 KST), plus three notable repos outside the chart — what each does, what changed this week, and what to watch out for. 10 android/skills · 9 ruflo · 8 VoiceStudio · 7 HyperFrames · 6 scientific-agent-skills · 5 Ghidra · 4 context-mode · 3 claude-plugins-official · 2 ai-job-search · 1 REA Extras: OpenCode (the harness runs locally, the model API only provides inference; Console security incident — rotate stored keys), vLLM (like Ollama but built for high-throughput serving; 0.32 restricts default CORS), cursor/plugins. Narration is an AI voice (ElevenLabs).

Top 10 GitHub Repos This Week | Oct 11 – Oct 17 | Agents Take Software Apart This week the agents learned to take compiled software apart: REA gained 239,923 stars in seven days, more than three times the other nine combined, with Ghidra and AnyPS5 on the same theme. ⏱️ TIMESTAMPS 0:00 — Intro 0:50 — #10 billion-context — Context compression for long agent sessions (1.8K⭐️) 2:49 — #9 claude-plugins-official — Anthropic's official Claude Code plugins (42K⭐️) 4:50 — #8 context-mode — Keeps tool output out of your context (30.6K⭐️) 6:58 — #7 hyperframes — Write HTML, render video, built for agents (66.1K⭐️) 9:03 — #6 ghidra — The NSA's free reverse engineering suite (90.4K⭐️) 11:04 — #5 diagram-design — Editorial diagrams from your agent (57.2K⭐️) 13:13 — #4 skills — Engineering skills for coding agents (295.3K⭐️) 15:13 — #3 AnyPS5 — PS5 executables relinked to run natively (42K⭐️) 17:37 — #2 artcraft — An IDE for AI images and video (32K⭐️) 20:09 — #1 rea — Reverse engineer anything with agents (310.8K⭐️) 23:18 — Recap & Trends 🔗 REPOS • #10 billion-context → ranxianglei/billion-context • #9 claude-plugins-official → anthropics/claude-plugins-official • #8 context-mode → mksglu/context-mode • #7 hyperframes → heygen-com/hyperframes • #6 ghidra → NationalSecurityAgency/ghidra • #5 diagram-design → cathrynlavery/diagram-design • #4 skills → mattpocock/skills • #3 AnyPS5 → boykopovar/AnyPS5 • #2 artcraft → storytold/artcraft • #1 rea → morluto/rea 👍 Like, 💬 Comment, 🔔 Subscribe, ⭐️ Star the repos! github #opensource #ai #coding #programming #trending #devtools #developer [ar] أفضل 10 مستودعات GitHub لهذا الأسبوع | 11 أكتوبر – 17 أكتوبر | الوكلاء يفككون البرمجيات تعلم العملاء هذا الأسبوع كيفية تحليل البرامج المُجمّعة: حصدت REA 239,923 نجمة في سبعة أيام، أي أكثر من ثلاثة أضعاف مجموع نجوم البرامج التسعة الأخرى، مع تشابه أداء Ghidra وAnyPS5. ... ⏱️ الطوابع الزمنية 0:00 — مقدمة 0:50 — #10 billion-context — ضغط السياق لجلسات الوكلاء الطويلة (1.8 ألف نجمة) 2:49 — #9 claude-plugins-official — إضافات Claude Code الرسمية من Anthropic (42 ألف نجمة) 4:50 — #8 context-mode — يبقي مخرجات الأداة خارج سياقك (30.6 ألف نجمة) 6:58 — #7 hyperframes — كتابة HTML، عرض فيديو، مصمم للوكلاء (66.1 ألف نجمة) 9:03 — #6 ghidra — مجموعة أدوات الهندسة العكسية المجانية من وكالة الأمن القومي (90.4 ألف نجمة) 11:04 — #5 diagram-design — مخططات تحريرية من وكيلك (57.2 ألف نجمة) 13:13 — #4 skills — مهارات هندسية لبرمجة الوكلاء (295.3 ألف نجمة) 15:13 — #3 AnyPS5 — إعادة ربط ملفات PS5 التنفيذية لتشغيلها بشكل أصلي (42 ألف نجمة) 17:37 — #2 artcraft — بيئة تطوير متكاملة (IDE) لصور وفيديوهات الذكاء الاصطناعي (32 ألف نجمة) 20:09 — #1 rea — الهندسة العكسية لأي شيء باستخدام الوكلاء (310.8 ألف نجمة) 23:18 — ملخص واتجاهات 🔗 المستودعات • #10 billion-context → ranxianglei/billion-context • #9 claude-plugins-official → anthropics/claude-plugins-official • #8 context-mode → mksglu/context-mode • #7 hyperframes → heygen-com/hyperframes • #6 ghidra → NationalSecurityAgency/ghidra • #5 تصميم المخططات → cathrynlavery/diagram-design • #4 المهارات → mattpocock/skills • #3 AnyPS5 → boykopovar/AnyPS5 • #2 Artcraft → storytold/artcraft • #1 rea → morluto/rea 👍 إعجاب، 💬 تعليق، 🔔 اشتراك، ⭐️ ضع نجمة على المستودعات! github #opensource #ai #coding #programming #trending #devtools #developer [bn] এই সপ্তাহের সেরা ১০টি গিটহাব রিপোজিটরি | ১১ অক্টোবর – ১৭ অক্টোবর | সফটওয়্যার বিশ্লেষণ করছে এআই এজ... এই সপ্তাহে এজেন্টরা কম্পাইল করা সফটওয়্যার বিশ্লেষণ করতে শিখেছে: REA সাত দিনে ২৩৯,৯২৩টি স্টার অর্জন করেছে, যা অন্য নয়টির সম্মিলিত সংখ্যার তিন গুণেরও বেশি; একই ধারায় ছিল Ghidra এবং AnyPS5। ⏱️ টাইমস্ট্যাম্প ০:০০ — ভূমিকা ০:৫০ — #১০ বিলিয়ন-কন্টেক্সট — দীর্ঘ এজেন্ট সেশনের জন্য কন্টেক্সট কম্প্রেশন (১.৮ হাজার⭐️) ২:৪৯ — #৯ ক্লড-প্লাগইনস-অফিশিয়াল — অ্যানথ্রোপিকের অফিশিয়াল ক্লড কোড প্লাগইন (৪২ হাজার⭐️) ৪:৫০ — #৮ কন্টেক্সট-মোড — টুলের আউটপুট আপনার কন্টেক্সটের বাইরে রাখে (৩০.৬ হাজার⭐️) ৬:৫৮ — #৭ হাইপারফ্রেমস — এইচটিএমএল লিখুন, ভিডিও রেন্ডার করুন, এজেন্টদের জন্য তৈরি (৬৬.১ হাজার⭐️) ৯:০৩ — #৬ ঘিড্রা — এনএসএ-র বিনামূল্যের রিভার্স ইঞ্জিনিয়ারিং স্যুট (৯০.৪ হাজার⭐️) ১১:০৪ — #৫ ডায়াগ্রাম-ডিজাইন — আপনার এজেন্ট থেকে এডিটোরিয়াল ডায়াগ্রাম (৫৭.২ হাজার⭐️) ১৩:১৩ — #৪ স্কিলস — এজেন্টদের কোডিং করার জন্য ইঞ্জিনিয়ারিং দক্ষতা (২৯৫.৩ হাজার⭐️) ১৫:১৩ — #৩ AnyPS5 — PS5 এক্সিকিউটেবলগুলোকে নেটিভভাবে চালানোর জন্য রিলিংক করা হয়েছে (৪২ হাজার⭐️) ১৭:৩৭ — #২ artcraft — AI ইমেজ এবং ভিডিওর জন্য একটি IDE (৩২ হাজার⭐️) ২০:০৯ — #১ rea — এজেন্ট ব্যবহার করে যেকোনো কিছু রিভার্স ইঞ্জিনিয়ার করুন (৩১০.৮ হাজার⭐️) ২৩:১৮ — সারসংক্ষেপ ও ট্রেন্ডস 🔗 রিপো • #১০ billion-context → ranxianglei/billion-context • #৯ claude-plugins-official → anthropics/claude-plugins-official • #৮ context-mode → mksglu/context-mode • #৭ hyperframes → heygen-com/hyperframes • #৬ ঘিড্রা → NationalSecurityAgency/ghidra • #৫ ডায়াগ্রাম-ডিজাইন → cathrynlavery/diagram-design • #৪ স্কিলস → mattpocock/skills • #৩ এনিপিএস৫ → boykopovar/AnyPS5 • #২ আর্টক্রাফট → storytold/artcraft • #১ রিয়া → morluto/rea 👍 লাইক, 💬 কমেন্ট, 🔔 সাবস্ক্রাইব, ⭐️ রিপোজিটরিগুলোতে স্টার দিন! github #opensource #ai #coding #programming #trending #devtools #developer [de-DE] Die Top 10 GitHub-Repos dieser Woche | 11. Okt. – 17. Okt. | KI-Agenten nehmen Software auseinander Diese Woche lernten die Agenten, kompilierte Software zu analysieren: REA erhielt in sieben Tagen 239.923 Sterne, mehr als dreimal so viele wie die anderen neun zusammen. Ghidra und AnyPS5 folgten demselben Trend. ⏱️ ZEITSTEMPEL 0:00 — Intro 0:50 — #10 billion-context — Kontextkomprimierung für lange Agentensitzungen (1,8K⭐️) 2:49 — #9 claude-plugins-official — Die offiziellen Claude-Code-Plugins von Anthropic (42K⭐️) 4:50 — #8 context-mode — Hält die Tool-Ausgabe vom Kontext fern (30,6K⭐️) 6:58 — #7 hyperframes — HTML schreiben, Video rendern, entwickelt für Agenten (66,1K⭐️) 9:03 — #6 ghidra — Die kostenlose Reverse-Engineering-Suite der NSA (90,4K⭐️) 11:04 — #5 diagram-design — Redaktionelle Diagramme direkt vom Agenten (57,2K⭐️) 13:13 — #4 skills — Programmierkenntnisse für Coding-Agenten (295,3K⭐️) 15:13 – Platz 3 AnyPS5 – PS5-Programme wurden neu verlinkt, um nativ zu laufen (42K⭐️) 17:37 – Platz 2 artcraft – Eine IDE für KI-Bilder und -Videos (32K⭐️) 20:09 – Platz 1 rea – Reverse Engineering mit Agenten (310,8K⭐️) 23:18 – Zusammenfassung & Trends 🔗 REPOS • Platz 10 billion-context → ranxianglei/billion-context • Platz 9 claude-plugins-official → anthropics/claude-plugins-official • Platz 8 context-mode → mksglu/context-mode • Platz 7 hyperframes → heygen-com/hyperframes • #6 ghidra → NationalSecurityAgency/ghidra • #5 diagram-design → cathrynlavery/diagram-design • #4 skills → mattpocock/skills • #3 AnyPS5 → boykopovar/AnyPS5 • #2 artcraft → storytold/artcraft • #1 rea → morluto/rea 👍 Like, 💬 Kommentar, 🔔 Abonnieren, ⭐️ Repos mit Stern markieren! github #opensource #ai #coding #programming #trending #devtools #developer [es-US] Los 10 mejores repositorios de GitHub de la semana | 11 de octubre – 17 de octubre | Los agentes ... Esta semana, los agentes aprendieron a analizar software compilado: REA obtuvo 239.923 estrellas en siete días, más del triple que las otras nueve juntas, con Ghidra y AnyPS5 en la misma línea temática. ⏱️ MARCAS DE TIEMPO 0:00 — Introducción 0:50 — #10 billion-context — Compresión de contexto para sesiones largas de agentes (1.8K⭐️) 2:49 — #9 claude-plugins-official — Plugins oficiales de Claude Code de Anthropic (42K⭐️) 4:50 — #8 context-mode — Mantiene la salida de la herramienta fuera de tu contexto (30.6K⭐️) 6:58 — #7 hyperframes — Escribe HTML, renderiza vídeo, diseñado para agentes (66.1K⭐️) 9:03 — #6 ghidra — El conjunto de herramientas gratuito de ingeniería inversa de la NSA (90.4K⭐️) 11:04 — #5 diagram-design — Diagramas editoriales desde tu agente (57.2K⭐️) 13:13 — #4 skills — Habilidades de ingeniería para agentes de programación (295.3K⭐️) 15:13 — #3 AnyPS5 — Ejecutables de PS5 re-enlazados para ejecutarse de forma nativa (42K⭐️) 17:37 — #2 artcraft — Un IDE para imágenes y vídeo con IA (32K⭐️) 20:09 — #1 rea — Ingeniería inversa de cualquier cosa con agentes (310.8K⭐️) 23:18 — Resumen y tendencias 🔗 REPOS • #10 billion-context → ranxianglei/billion-context • #9 claude-plugins-official → anthropics/claude-plugins-official • #8 context-mode → mksglu/context-mode • #7 hyperframes → heygen-com/hyperframes • #6 ghidra → NationalSecurityAgency/ghidra • #5 diagram-design → cathrynlavery/diagram-design • #4 skills → mattpocock/skills • #3 AnyPS5 → boykopovar/AnyPS5 • #2 artcraft → storytold/artcraft • #1 rea → morluto/rea 👍 Me gusta, 💬 Comenta, 🔔 Suscríbete, ⭐️ ¡Marca los repositorios como favoritos! github #opensource #ia #coding #programming #trending #devtools #developer [fr-FR] Top 10 des dépôts GitHub de la semaine | 11 oct. – 17 oct. | Des agents analysent les logiciels Cette semaine, les agents ont appris à analyser en profondeur les logiciels compilés : REA a récolté 239 923 étoiles en sept jours, soit plus de trois fois le score cumulé des neuf autres agents, Ghidra et AnyPS5 affichant le même profil. ⏱️ HORODATAGE 0:00 — Introduction 0:50 — #10 billion-context — Compression du contexte pour les longues sessions d'agent (1,8K⭐️) 2:49 — #9 claude-plugins-official — Plugins Claude Code officiels d'Anthropic (42K⭐️) 4:50 — #8 context-mode — Maintient la sortie de l'outil hors de votre contexte (30,6K⭐️) 6:58 — #7 hyperframes — Écrire du HTML, rendre de la vidéo, conçu pour les agents (66,1K⭐️) 9:03 — #6 ghidra — La suite de rétro-ingénierie gratuite de la NSA (90,4K⭐️) 11:04 — #5 diagram-design — Diagrammes éditoriaux de votre agent (57,2K⭐️) 13:13 — #4 skills — Compétences d'ingénierie pour le développement d'agents (295,3K⭐️) 15:13 — N° 3 AnyPS5 — Exécutables PS5 reliés pour une exécution native (42K⭐️) 17:37 — N° 2 artcraft — Un IDE pour l'IA dans les images et la vidéo (32K⭐️) 20:09 — N° 1 rea — Rétro-ingénierie de tout type d'application avec agents (310,8K⭐️) 23:18 — Récapitulatif et tendances 🔗 DÉPÔTS • N° 10 billion-context → ranxianglei/billion-context • N° 9 claude-plugins-official → anthropics/claude-plugins-official • N° 8 context-mode → mksglu/context-mode • N° 7 hyperframes → heygen-com/hyperframes • #6 ghidra → NationalSecurityAgency/ghidra • #5 diagram-design → cathrynlavery/diagram-design • #4 skills → mattpocock/skills • #3 AnyPS5 → boykopovar/AnyPS5 • #2 artcraft → storytold/artcraft • #1 rea → morluto/rea 👍 Aimez, 💬 Commentez, 🔔 Abonnez-vous, ⭐️ Mettez les dépôts en favoris ! github #opensource #ai #coding #programming #trending #devtools #developer [hi] इस हफ्ते के टॉप 10 GitHub रिपोज़ | 11 अक्टूबर – 17 अक्टूबर | एजेंट्स कर रहे सॉफ्टवेयर का विश्लेषण इस सप्ताह एजेंटों ने कंपाइल किए गए सॉफ़्टवेयर को अलग-अलग करके देखना सीखा: REA ने सात दिनों में 239,923 स्टार हासिल किए, जो कि इसी विषय पर आधारित अन्य नौ सॉफ़्टवेयर, Ghidra और AnyPS5 के कुल स्टार्स से तीन गुना से भी अधिक है। ⏱️ समय-सारणी 0:00 — परिचय 0:50 — #10 बिलियन-कॉन्टेक्स्ट — लंबे एजेंट सत्रों के लिए कॉन्टेक्स्ट कम्प्रेशन (1.8K⭐️) 2:49 — #9 क्लाउड-प्लगइन्स-ऑफिशियल — एंथ्रोपिक के आधिकारिक क्लाउड कोड प्लगइन्स (42K⭐️) 4:50 — #8 कॉन्टेक्स्ट-मोड — टूल आउटपुट को आपके कॉन्टेक्स्ट से बाहर रखता है (30.6K⭐️) 6:58 — #7 हाइपरफ्रेम्स — HTML लिखें, वीडियो रेंडर करें, एजेंटों के लिए निर्मित (66.1K⭐️) 9:03 — #6 घिड्रा — NSA का मुफ़्त रिवर्स इंजीनियरिंग सूट (90.4K⭐️) 11:04 — #5 डायग्राम-डिज़ाइन — आपके एजेंट से संपादकीय आरेख (57.2K⭐️) 13:13 — #4 स्किल्स — कोडिंग एजेंटों के लिए इंजीनियरिंग कौशल (295.3K⭐️) 15:13 — #3 AnyPS5 — PS5 निष्पादन योग्य फ़ाइलों को मूल रूप से चलाने के लिए पुनः लिंक किया गया (42K⭐️) 17:37 — #2 artcraft — AI छवियों और वीडियो के लिए एक IDE (32K⭐️) 20:09 — #1 rea — एजेंटों के साथ किसी भी चीज़ की रिवर्स इंजीनियरिंग करें (310.8K⭐️) 23:18 — सारांश और रुझान 🔗 REPOS • #10 billion-context → ranxianglei/billion-context • #9 claude-plugins-official → anthropics/claude-plugins-official • #8 context-mode → mksglu/context-mode • #7 हाइपरफ्रेम्स → heygen-com/hyperframes • #6 घिड्रा → NationalSecurityAgency/ghidra • #5 डायग्राम-डिज़ाइन → cathrynlavery/diagram-design • #4 स्किल्स → mattpocock/skills • #3 एनीपीएस5 → boykopovar/AnyPS5 • #2 आर्टक्राफ्ट → storytold/artcraft • #1 रिया → morluto/rea 👍 लाइक करें, 💬 कमेंट करें, 🔔 सब्सक्राइब करें, ⭐️ रिपॉजिटरी को स्टार दें! जीथब #ओपनसोर्स #एआई #कोडिंग #प्रोग्रामिंग #ट्रेंडिंग #डेवटूल्स #डेवलपर [id] 10 Repositori GitHub Teratas Minggu Ini | 11 Okt – 17 Okt | Agen Mengutak-atik Perangkat Lunak Minggu ini para agen belajar membongkar perangkat lunak yang telah dikompilasi: REA memperoleh 239.923 bintang dalam tujuh hari, lebih dari tiga kali lipat gabungan sembilan game lainnya, dengan Ghidra dan AnyPS5 juga memiliki tema yang sama. ⏱️ CAP WAKTU 0:00 — Pendahuluan 0:50 — #10 billion-context — Kompresi konteks untuk sesi agen yang panjang (1,8K⭐️) 2:49 — #9 claude-plugins-official — Plugin resmi Claude Code dari Anthropic (42K⭐️) 4:50 — #8 context-mode — Menjaga output alat agar tidak masuk ke dalam konteks Anda (30,6K⭐️) 6:58 — #7 hyperframes — Menulis HTML, merender video, dirancang untuk agen (66,1K⭐️) 9:03 — #6 ghidra — Perangkat lunak rekayasa balik gratis dari NSA (90,4K⭐️) 11:04 — #5 diagram-design — Diagram editorial dari agen Anda (57,2K⭐️) 13:13 — #4 skills — Keterampilan rekayasa untuk agen pengkodean (295.3K⭐️) 15:13 — #3 AnyPS5 — File eksekusi PS5 yang dihubungkan ulang agar berjalan secara native (42K⭐️) 17:37 — #2 artcraft — IDE untuk gambar dan video AI (32K⭐️) 20:09 — #1 rea — Rekayasa balik apa pun dengan agen (310.8K⭐️) 23:18 — Rekap & Tren 🔗 REPOS • #10 billion-context → ranxianglei/billion-context • #9 claude-plugins-official → anthropics/claude-plugins-official • #8 context-mode → mksglu/context-mode • #7 hyperframes → heygen-com/hyperframes • #6 ghidra → NationalSecurityAgency/ghidra • #5 diagram-design → cathrynlavery/diagram-design • #4 skills → mattpocock/skills • #3 AnyPS5 → boykopovar/AnyPS5 • #2 artcraft → storytold/artcraft • #1 rea → morluto/rea 👍 Suka, 💬 Komentar, 🔔 Berlangganan, ⭐️ Beri bintang pada repositori! github #opensource #ai #coding #programming #trending #devtools #developer [it] I 10 migliori repository GitHub della settimana | 11 ott – 17 ott | Gli agenti smontano i software Questa settimana gli agenti hanno imparato a smontare i software compilati: REA ha guadagnato 239.923 stelle in sette giorni, più del triplo rispetto agli altri nove messi insieme, con Ghidra e AnyPS5 sulla stessa linea. ⏱️ INDICE DEI CONTENUTI 0:00 — Introduzione 0:50 — #10 billion-context — Compressione del contesto per sessioni lunghe degli agenti (1.800⭐️) 2:49 — #9 claude-plugins-official — Plugin ufficiali di Anthropic per Claude Code (42.000⭐️) 4:50 — #8 context-mode — Mantiene l'output dello strumento fuori dal contesto (30.600⭐️) 6:58 — #7 hyperframes — Scrivi HTML, renderizza video, creato per gli agenti (66.100⭐️) 9:03 — #6 ghidra — La suite gratuita di reverse engineering della NSA (90.400⭐️) 11:04 — #5 diagram-design — Diagrammi editoriali dal tuo agente (57.200⭐️) 13:13 — #4 skills — Competenze ingegneristiche per agenti programmatori (295.300⭐️) 15:13 — #3 AnyPS5 — Eseguibili PS5 ricollegati per l'esecuzione nativa (42.000⭐️) 17:37 — #2 artcraft — Un IDE per immagini e video AI (32.000⭐️) 20:09 — #1 rea — Decodifica inversamente qualsiasi cosa con gli agenti (310.800⭐️) 23:18 — Riepilogo e tendenze 🔗 REPOS • #10 billion-context → ranxianglei/billion-context • #9 claude-plugins-official → anthropics/claude-plugins-official • #8 context-mode → mksglu/context-mode • #7 hyperframes → Italiano: heygen-com/hyperframes • #6 ghidra → NationalSecurityAgency/ghidra • #5 diagram-design → cathrynlavery/diagram-design • #4 skills → mattpocock/skills • #3 AnyPS5 → boykopovar/AnyPS5 • #2 artcraft → storytold/artcraft • #1 rea → morluto/rea 👍 Metti Mi piace, 💬 Commenta, 🔔 Iscriviti, ⭐️ Aggiungi i repository ai preferiti! github #opensource #ai #coding #programming #trending #devtools #developer [iw] 10 הריפוזיטורים המובילים ב-GitHub השבוע | 11 באוקטובר – 17 באוקטובר | סוכני בינה מלאכותית מפרקים ... השבוע למדו הסוכנים לפרק תוכנות שעברו קומפילציה: REA צברה 239,923 כוכבים בשבעה ימים, יותר מפי שלושה מתשעת האחרות גם יחד, כאשר Ghidra ו-AnyPS5 היו באותו נושא. ⏱️ חותמות זמן 0:00 — מבוא 0:50 — #10 billion-context — דחיסת הקשר עבור סשנים ארוכים של סוכנים (1.8K⭐️) 2:49 — #9 claude-plugins-official — התוספים הרשמיים של Claude Code של Anthropic (42K⭐️) 4:50 — #8 context-mode — שומר על פלט הכלי מחוץ להקשר שלך (30.6K⭐️) 6:58 — #7 hyperframes — כתיבת HTML, רינדור וידאו, בנוי עבור סוכנים (66.1K⭐️) 9:03 — #6 ghidra — חבילת ההנדסה הפוכה החינמית של ה-NSA (90.4K⭐️) 11:04 — #5 diagram-design — דיאגרמות עריכה מהסוכן שלך (57.2K⭐️) 13:13 — #4 skills — מיומנויות הנדסה עבור סוכני קידוד (295.3K⭐️) 15:13 — #3 AnyPS5 — קבצי הרצה של PS5 מקושרים מחדש להפעלה טבעית (42K⭐️) 17:37 — #2 artcraft — IDE לתמונות וסרטונים של בינה מלאכותית (32K⭐️) 20:09 — #1 rea — הנדסה הפוכה של כל דבר עם סוכנים (310.8K⭐️) 23:18 — סיכום ומגמות 🔗 REPOS • #10 billion-context → ranxianglei/billion-context • #9 claude-plugins-official → anthropics/claude-plugins-official • #8 context-mode → mksglu/context-mode • #7 hyperframes → heygen-com/hyperframes • #6 ghidra → NationalSecurityAgency/ghidra • #5 diagram-design → cathrynlavery/diagram-design • #4 מיומנויות → mattpocock/skills • #3 AnyPS5 → boykopovar/AnyPS5 • #2 artcraft → storytold/artcraft • #1 rea → morluto/rea 👍 לייק, 💬 תגובה, 🔔 הירשם, ⭐️ סמן בכוכב את המאגרים! github #קוד פתוח #בינה מלאכותית #קידוד #תכנות #טרנדי #כלי_פיתוח #מפתח [ja] 今週のGitHubリポジトリTOP10 | 10月11日～10月17日 | AIエージェントがソフトウェアを解体 今週、エージェントたちはコンパイル済みソフトウェアの解析方法を習得した。REAは7日間で239,923個の星を獲得し、同じテーマで活躍した他の9人のエージェントの合計の3倍以上となった。GhidraとAnyPS5も同様の成果を上げた。 ⏱️ タイムスタンプ 0:00 — はじめに 0:50 — #10 billion-context — 長時間のエージェントセッションのためのコンテキスト圧縮 (1.8K⭐️) 2:49 — #9 claude-plugins-official — Anthropic公式Claude Codeプラグイン (42K⭐️) 4:50 — #8 context-mode — ツールの出力をコンテキストから切り離します (30.6K⭐️) 6:58 — #7 hyperframes — HTMLの書き込み、ビデオのレンダリング、エージェント向けに構築 (66.1K⭐️) 9:03 — #6 ghidra — NSAの無料リバースエンジニアリングスイート (90.4K⭐️) 11:04 — #5 diagram-design — エージェントから編集用図を作成 (57.2K⭐️) 13:13 — #4 skills — エージェントをコーディングするためのエンジニアリングスキル (295.3K⭐️) 15:13 — #3 AnyPS5 — PS5ネイティブ実行用に再リンクされた実行ファイル (42K⭐️) 17:37 — #2 artcraft — AI画像と動画のためのIDE (32K⭐️) 20:09 — #1 rea — エージェントであらゆるものをリバースエンジニアリング (310.8K⭐️) 23:18 — まとめとトレンド 🔗 リポジトリ • #10 billion-context → ranxianglei/billion-context • #9 claude-plugins-official → anthropics/claude-plugins-official • #8 context-mode → mksglu/context-mode • #7 hyperframes → heygen-com/hyperframes • #6 ghidra → NationalSecurityAgency/ghidra • #5 diagram-design → cathrynlavery/diagram-design • #4 スキル → mattpocock/skills • #3 AnyPS5 → boykopovar/AnyPS5 • #2 アートクラフト → storytold/artcraft • #1 リア → morluto/rea 👍 いいね、💬 コメント、🔔 チャンネル登録、⭐️ スターをリポジトリに追加！ github #オープンソース #AI #コーディング #プログラミング #トレンド #開発ツール #開発者 [ko] 이번 주 GitHub 인기 저장소 TOP 10 | 10월 11일 – 10월 17일 | AI 에이전트의 소프트웨어 분석 이번 주 에이전트들은 컴파일된 소프트웨어를 분해하는 방법을 배웠습니다. REA는 7일 만에 239,923개의 별점을 획득했는데, 이는 같은 주제로 활동한 Ghidra와 AnyPS5를 포함한 다른 9개 프로그램의 별점 합계보다 3배 이상 많은 수치입니다. ⏱️ 타임스탬프 0:00 — 소개 0:50 — #10 billion-context — 장시간 에이전트 세션을 위한 컨텍스트 압축 (1.8K⭐️) 2:49 — #9 claude-plugins-official — Anthropic의 공식 Claude Code 플러그인 (42K⭐️) 4:50 — #8 context-mode — 도구 출력이 컨텍스트에 노출되지 않도록 함 (30.6K⭐️) 6:58 — #7 hyperframes — 에이전트를 위한 HTML 작성, 비디오 렌더링 (66.1K⭐️) 9:03 — #6 ghidra — NSA의 무료 리버스 엔지니어링 도구 모음 (90.4K⭐️) 11:04 — #5 diagram-design — 에이전트에서 편집 다이어그램 생성 (57.2K⭐️) 13:13 — #4 skills — 에이전트 코딩을 위한 엔지니어링 스킬 (295.3K⭐️) 15:13 — #3 AnyPS5 — PS5 실행 파일 네이티브 실행을 위해 다시 링크됨 (42K⭐️) 17:37 — #2 artcraft — AI 이미지 및 비디오용 IDE (32K⭐️) 20:09 — #1 rea — 에이전트를 사용하여 모든 것을 역설계 (310.8K⭐️) 23:18 — 요약 및 트렌드 🔗 저장소 • #10 billion-context → ranxianglei/billion-context • #9 claude-plugins-official → anthropics/claude-plugins-official • #8 context-mode → mksglu/context-mode • #7 hyperframes → heygen-com/hyperframes • #6 ghidra → NationalSecurityAgency/ghidra • #5 diagram-design → cathrynlavery/diagram-design • #4 스킬 → mattpocock/skills • #3 AnyPS5 → boykopovar/AnyPS5 • #2 아트크래프트 → storytold/artcraft • #1 rea → morluto/rea 👍 좋아요, 💬 댓글, 🔔 구독, ⭐️ 저장소에 별표 표시! github #opensource #ai #coding #programming #trending #devtools #developer [ml] ഈ ആഴ്ചയിലെ മികച്ച 10 GitHub റിപ്പോസിറ്ററികൾ | ഒക്ടോബർ 11 – ഒക്ടോബർ 17 | ഏജന്റുകൾ സോഫ്റ്റ്‌വെയറുകൾ... ഈ ആഴ്ച ഏജന്റുമാർ കംപൈൽ ചെയ്ത സോഫ്റ്റ്‌വെയർ വേർപെടുത്താൻ പഠിച്ചു: ഏഴ് ദിവസത്തിനുള്ളിൽ REA 239,923 നക്ഷത്രങ്ങൾ നേടി, മറ്റ് ഒമ്പത് നക്ഷത്രങ്ങളെ അപേക്ഷിച്ച് മൂന്നിരട്ടിയിലധികം, Ghidra ഉം AnyPS5 ഉം ഒരേ തീമിൽ. ⏱️ TIMESTAMPS 0:00 — ആമുഖം 0:50 — #10 ബില്യൺ-സന്ദർഭം — ദൈർഘ്യമേറിയ ഏജന്റ് സെഷനുകൾക്കുള്ള സന്ദർഭ കംപ്രഷൻ (1.8K⭐️) 2:49 — #9 ക്ലോഡ്-പ്ലഗിനുകൾ-ഔദ്യോഗികം — ആന്ത്രോപിക്കിന്റെ ഔദ്യോഗിക ക്ലോഡ് കോഡ് പ്ലഗിനുകൾ (42K⭐️) 4:50 — #8 സന്ദർഭ-മോഡ് — ടൂൾ ഔട്ട്‌പുട്ട് നിങ്ങളുടെ സന്ദർഭത്തിൽ നിന്ന് മാറ്റി നിർത്തുന്നു (30.6K⭐️) 6:58 — #7 ഹൈപ്പർഫ്രെയിമുകൾ — HTML എഴുതുക, വീഡിയോ റെൻഡർ ചെയ്യുക, ഏജന്റുകൾക്കായി നിർമ്മിച്ചത് (66.1K⭐️) 9:03 — #6 ഘിദ്ര — NSA-യുടെ സൗജന്യ റിവേഴ്സ് എഞ്ചിനീയറിംഗ് സ്യൂട്ട് (90.4K⭐️) 11:04 — #5 ഡയഗ്രം-ഡിസൈൻ — നിങ്ങളുടെ ഏജന്റിൽ നിന്നുള്ള എഡിറ്റോറിയൽ ഡയഗ്രമുകൾ (57.2K⭐️) 13:13 — #4 കഴിവുകൾ — കോഡിംഗ് ഏജന്റുകൾക്കുള്ള എഞ്ചിനീയറിംഗ് കഴിവുകൾ (295.3K⭐️) 15:13 — #3 AnyPS5 — PS5 എക്സിക്യൂട്ടബിളുകൾ നേറ്റീവ് ആയി പ്രവർത്തിപ്പിക്കാൻ വീണ്ടും ലിങ്ക് ചെയ്‌തു (42K⭐️) 17:37 — #2 ആർട്ട്‌ക്രാഫ്റ്റ് — AI ഇമേജുകൾക്കും വീഡിയോയ്ക്കുമുള്ള ഒരു IDE (32K⭐️) 20:09 — #1 rea — ഏജന്റുകൾ ഉപയോഗിച്ച് എന്തും റിവേഴ്‌സ് എഞ്ചിനീയർ ചെയ്യുക (310.8K⭐️) 23:18 — റീക്യാപ്പ് & ട്രെൻഡുകൾ 🔗 REPOS • #10 ബില്യൺ-സന്ദർഭം → ranxianglei/billion-%E0%B4%B8%E0%B4%A8%E0%B5%8D%E0%B4%A6%E0%B5%BC%E0%B4%AD%E0%B4%82 • #9 claude-plugins-official → anthropics/claude-plugins-official • #8 context-mode → mksglu/context-mode • #7 hyperframes → heygen-com/hyperframes • #6 ghidra → NationalSecurityAgency/ghidra • #5 diagram-design → cathrynlavery/diagram-design • #4 കഴിവുകൾ → mattpocock/skills • #3 AnyPS5 → boykopovar/AnyPS5 • #2 ആർട്ട്‌ക്രാഫ്റ്റ് → storytold/artcraft • #1 rea → morluto/rea 👍 ലൈക്ക് ചെയ്യുക, 💬 കമന്റ് ചെയ്യുക, 🔔 സബ്‌സ്‌ക്രൈബ് ചെയ്യുക, ⭐️ റിപ്പോകളിൽ സ്റ്റാർ ചെയ്യുക! ഗിത്തബ് #ഓപ്പൺസോഴ്‌സ് #എഐ #കോഡിംഗ് #പ്രോഗ്രാമിംഗ് #ട്രെൻഡിംഗ് #ഡെവലപ്പർ #ഡെവലപ്പർ [nl-NL] Top 10 GitHub Repos van deze week | 11 okt – 17 okt | Agents nemen software onder handen Deze week leerden de agenten hoe ze gecompileerde software moesten ontleden: REA verdiende 239.923 sterren in zeven dagen, meer dan drie keer zoveel als de andere negen samen, met Ghidra en AnyPS5 die hetzelfde thema volgden. ⏱️ TIJDSTEMPELS 0:00 — Intro 0:50 — #10 billion-context — Contextcompressie voor lange agentsessies (1,8K⭐️) 2:49 — #9 claude-plugins-official — De officiële Claude Code-plugins van Anthropic (42K⭐️) 4:50 — #8 context-mode — Houdt de uitvoer van tools buiten je context (30,6K⭐️) 6:58 — #7 hyperframes — Schrijf HTML, render video, gebouwd voor agents (66,1K⭐️) 9:03 — #6 ghidra — De gratis reverse engineering suite van de NSA (90,4K⭐️) 11:04 — #5 diagram-design — Redactionele diagrammen vanuit je agent (57,2K⭐️) 13:13 — #4 skills — Technische vaardigheden voor coderende agents (295,3K⭐️) 15:13 — #3 AnyPS5 — PS5-executables opnieuw gekoppeld om native te draaien (42K⭐️) 17:37 — #2 artcraft — Een IDE voor AI-afbeeldingen en -video (32K⭐️) 20:09 — #1 rea — Reverse-engineer alles met agents (310,8K⭐️) 23:18 — Samenvatting & Trends 🔗 REPOS • #10 billion-context → ranxianglei/billion-context • #9 claude-plugins-official → anthropics/claude-plugins-official • #8 context-mode → mksglu/context-mode • #7 hyperframes → heygen-com/hyperframes • #6 ghidra → NationalSecurityAgency/ghidra • #5 diagram-design → cathrynlavery/diagram-design • #4 skills → mattpocock/skills • #3 AnyPS5 → boykopovar/AnyPS5 • #2 artcraft → storytold/artcraft • #1 rea → morluto/rea 👍 Vind ik leuk, 💬 Reageer, 🔔 Abonneer, ⭐️ Geef de repositories een ster! github #opensource #ai #coding #programming #trending #devtools #developer [pa] ਇਸ ਹਫ਼ਤੇ ਦੇ ਚੋਟੀ ਦੇ 10 GitHub ਰੈਪੋ | 11 ਅਕਤੂਬਰ - 17 ਅਕਤੂਬਰ | ਏਜੰਟਾਂ ਨੇ ਸਾਫਟਵੇਅਰ ਦੇ ਪੁਰਜ਼ੇ ਵੱਖ ਕੀਤੇ ਇਸ ਹਫ਼ਤੇ ਏਜੰਟਾਂ ਨੇ ਕੰਪਾਇਲ ਕੀਤੇ ਸੌਫਟਵੇਅਰ ਨੂੰ ਵੱਖ ਕਰਨਾ ਸਿੱਖਿਆ: REA ਨੇ ਸੱਤ ਦਿਨਾਂ ਵਿੱਚ 239,923 ਸਟਾਰ ਪ੍ਰਾਪਤ ਕੀਤੇ, ਜੋ ਕਿ ਬਾਕੀ ਨੌਂ ਜੋੜਿਆਂ ਨਾਲੋਂ ਤਿੰਨ ਗੁਣਾ ਵੱਧ ਹਨ, ਘਿਦਰਾ ਅਤੇ AnyPS5 ਇੱਕੋ ਥੀਮ 'ਤੇ ਹਨ। ⏱️ ਟਾਈਮਸਟੈਂਪਸ 0:00 — ਜਾਣ-ਪਛਾਣ 0:50 — #10 ਬਿਲੀਅਨ-ਸੰਦਰਭ — ਲੰਬੇ ਏਜੰਟ ਸੈਸ਼ਨਾਂ ਲਈ ਸੰਦਰਭ ਸੰਕੁਚਨ (1.8K⭐️) 2:49 — #9 ਕਲਾਉਡ-ਪਲੱਗਇਨ-ਅਧਿਕਾਰਤ — ਐਂਥ੍ਰੋਪਿਕ ਦਾ ਅਧਿਕਾਰਤ ਕਲਾਉਡ ਕੋਡ ਪਲੱਗਇਨ (42K⭐️) 4:50 — #8 ਸੰਦਰਭ-ਮੋਡ — ਟੂਲ ਆਉਟਪੁੱਟ ਨੂੰ ਤੁਹਾਡੇ ਸੰਦਰਭ ਤੋਂ ਬਾਹਰ ਰੱਖਦਾ ਹੈ (30.6K⭐️) 6:58 — #7 ਹਾਈਪਰਫ੍ਰੇਮ — HTML ਲਿਖੋ, ਵੀਡੀਓ ਰੈਂਡਰ ਕਰੋ, ਏਜੰਟਾਂ ਲਈ ਬਣਾਇਆ ਗਿਆ (66.1K⭐️) 9:03 — #6 ਘਿਦਰਾ — NSA ਦਾ ਮੁਫ਼ਤ ਰਿਵਰਸ ਇੰਜੀਨੀਅਰਿੰਗ ਸੂਟ (90.4K⭐️) 11:04 — #5 ਡਾਇਗ੍ਰਾਮ-ਡਿਜ਼ਾਈਨ — ਤੁਹਾਡੇ ਏਜੰਟ ਤੋਂ ਸੰਪਾਦਕੀ ਡਾਇਗ੍ਰਾਮ (57.2K⭐️) 13:13 — #4 ਹੁਨਰ — ਕੋਡਿੰਗ ਏਜੰਟਾਂ ਲਈ ਇੰਜੀਨੀਅਰਿੰਗ ਹੁਨਰ (295.3K⭐️) 15:13 — #3 ਕੋਈ ਵੀPS5 — PS5 ਐਗਜ਼ੀਕਿਊਟੇਬਲ ਨੂੰ ਨੇਟਿਵਲੀ ਚਲਾਉਣ ਲਈ ਦੁਬਾਰਾ ਲਿੰਕ ਕੀਤਾ ਗਿਆ (42K⭐️) 17:37 — #2 ਆਰਟਕ੍ਰਾਫਟ — AI ਚਿੱਤਰਾਂ ਅਤੇ ਵੀਡੀਓ ਲਈ ਇੱਕ IDE (32K⭐️) 20:09 — #1 rea — ਏਜੰਟਾਂ ਨਾਲ ਕੁਝ ਵੀ ਰਿਵਰਸ ਇੰਜੀਨੀਅਰ (310.8K⭐️) 23:18 — ਰੀਕੈਪ ਅਤੇ ਰੁਝਾਨ 🔗 REPOS • #10 ਬਿਲੀਅਨ-ਸੰਦਰਭ → ranxianglei/billion-context • #9 claude-plugins-official → anthropics/claude-plugins-official • #8 ਸੰਦਰਭ-ਮੋਡ → mksglu/context-mode • #7 ਹਾਈਪਰਫ੍ਰੇਮ → heygen-com/hyperframes • #6 ghidra → NationalSecurityAgency/ghidra • #5 ਡਾਇਗ੍ਰਾਮ-ਡਿਜ਼ਾਈਨ → cathrynlavery/diagram-design • #4 ਹੁਨਰ → mattpocock/skills • #3 AnyPS5 → boykopovar/AnyPS5 • #2 artcraft → storytold/artcraft • #1 rea → morluto/rea 👍 ਪਸੰਦ ਕਰੋ, 💬 ਟਿੱਪਣੀ ਕਰੋ, 🔔 ਸਬਸਕ੍ਰਾਈਬ ਕਰੋ, ⭐️ ਰਿਪੋਜ਼ ਨੂੰ ਸਟਾਰ ਕਰੋ! github #opensource #ai #coding #programming #trending #devtools #developer [pl] Top 10 repozytoriów GitHub w tym tygodniu | 11 – 17 października | Agenci rozkładają oprogramowan... W tym tygodniu agenci nauczyli się rozkładać skompilowane oprogramowanie na czynniki pierwsze: REA zdobyła 239 923 gwiazdki w ciągu siedmiu dni, ponad trzy razy więcej niż pozostałe dziewięć razem wzięte, a Ghidra i AnyPS5 poruszały ten sam temat. ⏱️ ZNACZNIKI CZASU 0:00 — Wprowadzenie 0:50 — #10 billion-context — Kompresja kontekstu dla długich sesji agentów (1,8 tys.⭐️) 2:49 — #9 claude-plugins-official — Oficjalne wtyczki Claude Code firmy Anthropic (42 tys.⭐️) 4:50 — #8 context-mode — Utrzymuje dane wyjściowe narzędzia poza kontekstem (30,6 tys.⭐️) 6:58 — #7 hyperframes — Pisanie kodu HTML, renderowanie wideo, stworzone dla agentów (66,1 tys.⭐️) 9:03 — #6 ghidra — Darmowy pakiet NSA do inżynierii wstecznej (90,4 tys.⭐️) 11:04 — #5 diagram-design — Diagramy redakcyjne z agenta (57,2 tys.⭐️) 13:13 — #4 skills — Umiejętności inżynierskie dla agentów kodujących (295,3 tys.⭐️) 15:13 — #3 AnyPS5 — Pliki wykonywalne PS5 ponownie połączone, aby działały natywnie (42 tys.⭐️) 17:37 — #2 artcraft — IDE do obrazów i wideo AI (32 tys.⭐️) 20:09 — #1 rea — Inżynieria wsteczna wszystkiego za pomocą agentów (310,8 tys.⭐️) 23:18 — Podsumowanie i trendy 🔗 REPOZYTY • #10 billion-context → ranxianglei/billion-context • #9 claude-plugins-official → anthropics/claude-plugins-official • #8 context-mode → mksglu/context-mode • #7 hyperframes → heygen-com/hyperframes • #6 ghidra → NationalSecurityAgency/ghidra • #5 diagram-design → cathrynlavery/diagram-design • #4 umiejętności → mattpocock/skills • #3 AnyPS5 → boykopovar/AnyPS5 • #2 artcraft → storytold/artcraft • #1 rea → morluto/rea 👍 Polub, 💬 Skomentuj, 🔔 Subskrybuj, ⭐️ Oznacz repozytoria gwiazdką! github #opensource #sztuczna inteligencja #kodowanie #programowanie #trendy #narzędzia_programistyczne #programista [pt-BR] Top 10 Repositórios do GitHub desta semana | 11 de out – 17 de out | Agentes Desmontam Softwares Esta semana, os agentes aprenderam a desmontar softwares compilados: REA ganhou 239.923 estrelas em sete dias, mais de três vezes o total dos outros nove combinados, com Ghidra e AnyPS5 na mesma categoria. ⏱️ MARCAÇÕES DE TEMPO 0:00 — Introdução 0:50 — #10 billion-context — Compressão de contexto para sessões longas de agentes (1,8 mil ⭐️) 2:49 — #9 claude-plugins-official — Plugins oficiais do Claude Code da Anthropic (42 mil ⭐️) 4:50 — #8 context-mode — Mantém a saída da ferramenta fora do seu contexto (30,6 mil ⭐️) 6:58 — #7 hyperframes — Escreva HTML, renderize vídeo, desenvolvido para agentes (66,1 mil ⭐️) 9:03 — #6 ghidra — O pacote gratuito de engenharia reversa da NSA (90,4 mil ⭐️) 11:04 — #5 diagram-design — Diagramas editoriais do seu agente (57,2 mil ⭐️) 13:13 — #4 skills — Habilidades de engenharia para agentes de codificação (295,3 mil⭐️) 15:13 — #3 AnyPS5 — Executáveis ​​do PS5 religados para rodar nativamente (42 mil⭐️) 17:37 — #2 artcraft — Uma IDE para imagens e vídeos com IA (32 mil⭐️) 20:09 — #1 rea — Faça engenharia reversa de qualquer coisa com agentes (310,8 mil⭐️) 23:18 — Resumo e Tendências 🔗 REPOS • #10 billion-context → ranxianglei/billion-context • #9 claude-plugins-official → anthropics/claude-plugins-official • #8 context-mode → mksglu/context-mode • #7 hyperframes → heygen-com/hyperframes • #6 ghidra → NationalSecurityAgency/ghidra • #5 diagram-design → cathrynlavery/diagram-design • #4 skills → mattpocock/skills • #3 AnyPS5 → boykopovar/AnyPS5 • #2 artcraft → storytold/artcraft • #1 rea → morluto/rea 👍 Curtir, 💬 Comentar, 🔔 Inscrever-se, ⭐️ Favoritar os repositórios! github #opensource #ai #coding #programming #trending #devtools #developer [ru] Топ-10 репозиториев GitHub за неделю | 11–17 октября | Агенты разбирают софт по винтикам На этой неделе агенты научились разбирать скомпилированное программное обеспечение: REA за семь дней набрала 239 923 звезды, что более чем в три раза превышает суммарное количество звезд за остальные девять дней, как и Ghidra и AnyPS5. ⏱️ ВРЕМЕННЫЕ МЕТКИ 0:00 — Вступление 0:50 — #10 billion-context — Сжатие контекста для длительных сеансов работы агентов (1,8 тыс.⭐️) 2:49 — #9 claude-plugins-official — Официальные плагины Claude Code от Anthropic (42 тыс.⭐️) 4:50 — #8 context-mode — Вывод инструмента вне контекста (30,6 тыс.⭐️) 6:58 — #7 hyperframes — Написание HTML, рендеринг видео, создано для агентов (66,1 тыс.⭐️) 9:03 — #6 ghidra — Бесплатный набор инструментов для обратного проектирования от АНБ (90,4 тыс.⭐️) 11:04 — #5 diagram-design — Редактирование диаграмм от вашего агента (57,2 тыс.⭐️) 13:13 — #4 skills — Инженерные навыки для кодирования агентов (295,3 тыс.⭐️) 15:13 — #3 AnyPS5 — Исполняемые файлы PS5 перелинкованы для нативного запуска (42 тыс.⭐️) 17:37 — #2 artcraft — IDE для обработки изображений и видео с помощью ИИ (32 тыс.⭐️) 20:09 — #1 rea — Обратное проектирование чего угодно с помощью агентов (310,8 тыс.⭐️) 23:18 — Краткий обзор и тренды 🔗 РЕПОЗИЦИИ • #10 billion-context → ranxianglei/billion-context • #9 claude-plugins-official → anthropics/claude-plugins-official • #8 context-mode → mksglu/context-mode • #7 hyperframes → heygen-com/hyperframes • #6 ghidra → NationalSecurityAgency/ghidra • #5 diagram-design → cathrynlavery/diagram-design • #4 skills → mattpocock/skills • #3 AnyPS5 → boykopovar/AnyPS5 • #2 artcraft → storytold/artcraft • #1 rea → morluto/rea 👍 Ставьте лайки, 💬 оставляйте комментарии, 🔔 подписывайтесь, ⭐️ отмечайте репозитории звездочкой! github #opensource #ai #coding #programming #trending #devtools #developer [ta] இந்த வாரத்தின் சிறந்த 10 GitHub களஞ்சியங்கள் | அக்டோபர் 11 – அக்டோபர் 17 | மென்பொருட்களைப் பிரிக்... இந்த வாரம், தொகுக்கப்பட்ட மென்பொருளைப் பிரித்துப் பார்க்க முகவர்கள் கற்றுக்கொண்டனர்: REA ஏழு நாட்களில் 239,923 ஸ்டார்களைப் பெற்றது, இது மற்ற ஒன்பது மென்பொருள்களின் மொத்தப் புள்ளிகளை விட மூன்று மடங்குக்கும் அதிகமாகும்; இதே கருப்பொருளில் Ghidra மற்றும் AnyPS5 ஆகியவையும் இருந்தன. ⏱️ நேரக்குறிப்புகள் 0:00 — அறிமுகம் 0:50 — #10 பில்லியன்-கான்டெக்ஸ்ட் — நீண்ட ஏஜென்ட் அமர்வுகளுக்கான கான்டெக்ஸ்ட் சுருக்கம் (1.8K⭐️) 2:49 — #9 கிளாட்-பிளகின்கள்-அஃபிஷியல் — ஆந்த்ரோபிக்கின் அதிகாரப்பூர்வ கிளாட் கோட் பிளகின்கள் (42K⭐️) 4:50 — #8 கான்டெக்ஸ்ட்-மோட் — கருவி வெளியீட்டை உங்கள் கான்டெக்ஸ்டிலிருந்து வெளியே வைக்கிறது (30.6K⭐️) 6:58 — #7 ஹைப்பர்ஃபிரேம்ஸ் — HTML எழுதவும், வீடியோவை ரெண்டர் செய்யவும், ஏஜென்ட்களுக்காக உருவாக்கப்பட்டது (66.1K⭐️) 9:03 — #6 கிட்ரா — NSA-யின் இலவச ரிவர்ஸ் இன்ஜினியரிங் தொகுப்பு (90.4K⭐️) 11:04 — #5 டயக்ராம்-டிசைன் — உங்கள் ஏஜென்ட்டிலிருந்து எடிட்டோரியல் வரைபடங்கள் (57.2K⭐️) 13:13 — #4 ஸ்கில்ஸ் — கோடிங் ஏஜென்ட்களுக்கான இன்ஜினியரிங் திறன்கள் (295.3K⭐️) 15:13 — #3 AnyPS5 — PS5 இயங்கு கோப்புகள் இயல்பாக இயங்குமாறு மீண்டும் இணைக்கப்பட்டுள்ளன (42K⭐️) 17:37 — #2 artcraft — AI படங்கள் மற்றும் வீடியோவிற்கான ஒரு IDE (32K⭐️) 20:09 — #1 rea — ஏஜென்ட்களைக் கொண்டு எதையும் ரிவர்ஸ் இன்ஜினியரிங் செய்தல் (310.8K⭐️) 23:18 — மீள்பார்வை & போக்குகள் 🔗 ரெப்போக்கள் • #10 billion-context → ranxianglei/billion-context • #9 claude-plugins-official → anthropics/claude-plugins-official • #8 context-mode → mksglu/context-mode • #7 hyperframes → heygen-com/hyperframes • #6 ghidra → NationalSecurityAgency/ghidra • #5 diagram-design → cathrynlavery/diagram-design • #4 skills → mattpocock/skills • #3 AnyPS5 → boykopovar/AnyPS5 • #2 artcraft → storytold/artcraft • #1 rea → morluto/rea 👍 லைக், 💬 கமெண்ட், 🔔 சப்ஸ்கிரைப், ⭐️ ரெப்போக்களை ஸ்டார் செய்யுங்கள்! github #opensource #ai #coding #programming #trending #devtools #developer [te] ఈ వారపు టాప్ 10 GitHub రిపోలు | అక్టోబర్ 11 - అక్టోబర్ 17 | సాఫ్ట్‌వేర్‌ను విడగొడుతున్న AI ఏజెంట్లు ఈ వారం ఏజెంట్లు కంపైల్ చేసిన సాఫ్ట్‌వేర్‌ను విడదీయడం నేర్చుకున్నారు: REA ఏడు రోజుల్లో 239,923 స్టార్లను సంపాదించింది, ఇది మిగిలిన తొమ్మిదింటినీ కలిపి సంపాదించిన దానికంటే మూడు రెట్లు ఎక్కువ. ఇదే కోవలో ఘిద్రా మరియు AnyPS5 కూడా నిలిచాయి. ⏱️ టైమ్‌స్టాంప్‌లు 0:00 — పరిచయం 0:50 — #10 బిలియన్-కాంటెక్స్ట్ — సుదీర్ఘ ఏజెంట్ సెషన్‌ల కోసం కాంటెక్స్ట్ కంప్రెషన్ (1.8K⭐️) 2:49 — #9 క్లాడ్-ప్లగిన్స్-అఫీషియల్ — ఆంత్రోపిక్ వారి అధికారిక క్లాడ్ కోడ్ ప్లగిన్‌లు (42K⭐️) 4:50 — #8 కాంటెక్స్ట్-మోడ్ — టూల్ అవుట్‌పుట్‌ను మీ కాంటెక్స్ట్ నుండి వేరుగా ఉంచుతుంది (30.6K⭐️) 6:58 — #7 హైపర్‌ఫ్రేమ్స్ — HTML రాయండి, వీడియోను రెండర్ చేయండి, ఏజెంట్ల కోసం రూపొందించబడింది (66.1K⭐️) 9:03 — #6 ఘిద్రా — NSA వారి ఉచిత రివర్స్ ఇంజనీరింగ్ సూట్ (90.4K⭐️) 11:04 — #5 డయాగ్రామ్-డిజైన్ — మీ ఏజెంట్ నుండి ఎడిటోరియల్ డయాగ్రామ్‌లు (57.2K⭐️) 13:13 — #4 స్కిల్స్ — కోడింగ్ ఏజెంట్ల కోసం ఇంజనీరింగ్ నైపుణ్యాలు (295.3K⭐️) 15:13 — #3 AnyPS5 — PS5 ఎగ్జిక్యూటబుల్స్ నేటివ్‌గా రన్ చేయడానికి రీలింక్ చేయబడ్డాయి (42K⭐️) 17:37 — #2 artcraft — AI చిత్రాలు మరియు వీడియో కోసం ఒక IDE (32K⭐️) 20:09 — #1 rea — ఏజెంట్లతో దేనినైనా రివర్స్ ఇంజనీర్ చేయండి (310.8K⭐️) 23:18 — రీక్యాప్ & ట్రెండ్స్ 🔗 REPOS • #10 billion-context → ranxianglei/billion-context • #9 claude-plugins-official → anthropics/claude-plugins-official • #8 context-mode → mksglu/context-mode • #7 hyperframes → heygen-com/hyperframes • #6 ఘిద్రా → NationalSecurityAgency/ghidra • #5 డయాగ్రామ్-డిజైన్ → cathrynlavery/diagram-design • #4 స్కిల్స్ → mattpocock/skills • #3 ఎనీపీఎస్5 → boykopovar/AnyPS5 • #2 ఆర్ట్‌క్రాఫ్ట్ → storytold/artcraft • #1 రియా → morluto/rea 👍 లైక్ చేయండి, 💬 కామెంట్ చేయండి, 🔔 సబ్‌స్క్రైబ్ చేయండి, ⭐️ రిపోలకు స్టార్ ఇవ్వండి! github #ఓపెన్‌సోర్స్ #ఏఐ #కోడింగ్ #ప్రోగ్రామింగ్ #ట్రెండింగ్ #డెవ్‌టూల్స్ #డెవలపర్ [uk] Топ-10 репозиторіїв GitHub цього тижня | 11 жовтня – 17 жовтня | Агенти розбирають ПЗ на частини Цього тижня агенти навчилися розбирати скомпільоване програмне забезпечення: REA отримала 239 923 зірки за сім днів, що більш ніж утричі більше, ніж інші дев'ять разом узятих, причому Ghidra та AnyPS5 були на ту саму тему. ⏱️ ЧАСОВІ ПОМІТКИ 0:00 — Вступ 0:50 — #10 billion-context — Стиснення контексту для довгих сеансів агента (1.8K⭐️) 2:49 — #9 claude-plugins-official — Офіційні плагіни Claude Code від Anthropic (42K⭐️) 4:50 — #8 context-mode — Виводить вивід інструменту поза контекстом (30.6K⭐️) 6:58 — #7 hyperframes — Напишіть HTML, візуалізуйте відео, створюйте для агентів (66.1K⭐️) 9:03 — #6 ghidra — Безкоштовний пакет зворотного проектування NSA (90.4K⭐️) 11:04 — #5 diagram-design — Редакційні діаграми від вашого агента (57.2K⭐️) 13:13 — #4 skills — Інженерні навички для агентів кодування (295.3K⭐️) 15:13 — #3 AnyPS5 — Виконувані файли PS5 перелінковані для роботи в нативному режимі (42 тис.⭐️) 17:37 — #2 artcraft — IDE для зображень та відео зі штучним інтелектом (32 тис.⭐️) 20:09 — #1 rea — Зворотне проектування будь-чого за допомогою агентів (310,8 тис.⭐️) 23:18 — Підсумки та тенденції 🔗 РЕПОРТАЖІЇ • #10 billion-context → ranxianglei/billion-context • #9 claude-plugins-official → anthropics/claude-plugins-official • #8 context-mode → mksglu/context-mode • #7 hyperframes → heygen-com/hyperframes • #6 ghidra → NationalSecurityAgency/ghidra • #5 diagram-design → cathrynlavery/diagram-design • #4 навички → mattpocock/skills • #3 AnyPS5 → boykopovar/AnyPS5 • #2 мистецтво → storytold/artcraft • #1 реа → morluto/rea 👍 Вподобайкайте, 💬 Коментуйте, 🔔 Підпишіться, ⭐️ Позначте репозиторії зірками! github #відкритийвихіднийкод #штучнийінтелект #кодування #програмування #тренди #інструментирозробки #розробник

Claude Code just got an official plugin directory that hit 41k stars in days Anthropic is now maintaining a public marketplace of high-quality plugins so you can extend Claude Code without hunting random repos anthropics/claude-plugins-official

今日のGitHubトレンド anthropics/claude-plugins-official Claude Code向けの高品質なプラグインをまとめたキュレーションディレクトリです。 Anthropic社製および外部パートナー製のプラグインを提供し、ユーザーが簡単にインストールして機能を拡張できる環境を提供することを目的としています。

Before you ship, run Claude’s official security plugin. Large repos may take ~30 minutes and use tokens without visible output. Still better than saying “we should’ve checked” after a security incident. ⁠ github.com/anthropics/claude-plugins-official/tre… ⁠