Microsoft and UC Berkeley just open-sourced TokenEconomizer (microsoft/TokenEconomizer) to fix token bloat in Claude Code and OpenAI Codex.
Tested across 277 SWE-Bench repositories, it cuts total coding agent cost by 18.4% to 30.9% using repo micro-skills memory, dynamic MCP tool pruning, and a dead-end loop guard.
3 likes · 416 views
coding agents waste tokens relearning the same repo every session. which test command works, which tools never get used.
microsoft + berkeley open-sourced a claude code / codex plugin that learns from past sessions. 18-31% cheaper on swe-bench.
microsoft/TokenEconomizer t.co
18 likes · 560 views
microsoft open sourced tokeneconomizer, a plugin that makes claude code, codex and copilot cli cheaper 💸
remembers repo facts, drops unused tools, breaks failure loops
they claim 18-31% lower cost on swe-bench. agents need memory, not bigger models
3 likes
microsoft open sourced tokeneconomizer, a plugin that makes claude code, codex and copilot cli cheaper 💸
remembers repo facts, drops unused tools, breaks failure loops
they claim 18-31% lower cost on swe-bench. agents need memory, not bigger models
microsoft/TokenEconomizer
3 likes · 202 views
Microsoft ➡️ "TokenEconomizer: Experience-Driven Black-Box Cost Control for Coding Agents. Our measurement on SWE-Bench tasks: 18.4–30.9% lower mean cost across five agent–model combinations, cheaper in all 30 repository × combination cells, and Pass@1 up in 4 of 5." ➡️ Good for budget? :P
microsoft/TokenEconomizer
Community Overview: lifehubber.com
Resources are shared for discovery and are not independently vetted—please do your own due diligence.
New resources are added regularly — feel free to join the sub for updates.
Full searchable archive of all resources posted so far on our community site, LifeHubber: lifehubber.com 300+ open-ish AI models, agents, tools, datasets, and related resources, with filtering and sorting.
3 points
Microsoft ➡️ "TokenEconomizer: Experience-Driven Black-Box Cost Control for Coding Agents. Our measurement on SWE-Bench tasks: 18.4–30.9% lower mean cost across five agent–model combinations, cheaper in all 30 repository × combination cells, and Pass@1 up in 4 of 5." ➡️ Good for budget? :P
microsoft/TokenEconomizer
Community Overview: lifehubber.com
Resources are shared for discovery and are not independently vetted—please do your own due diligence.
New resources are added regularly — feel free to join the sub for updates.
Full searchable archive of all resources posted so far on our community site, LifeHubber: lifehubber.com 300+ open-ish AI models, agents, tools, datasets, and related resources, with filtering and sorting.
11 points
Microsoft's TokenEconomizer cuts coding agent costs up to 30.9%, and it does it by editing your config files
The short version
Microsoft Research, UC Berkeley and Microsoft Security published TokenEconomizer last week (v0.1.0, October 26, Apache 2.0). It is a plugin for Claude Code, Codex, the GitHub Copilot CLI and Pi that cuts what your agent costs to run, without touching model weights or agent code. On 277 SWE-Bench tasks across six repositories it lowered mean cost per task by 18.4 to 30.9 percent across five agent-model combinations, it came out cheaper in every one of the 30 repository-by-combination cells, and Pass@1 went up in four of the five combinations (paper).
I spend my days on agent config files, so the interesting part for me is not the percentage. It is how the thing works: it learns from your finished sessions, then edits your agent's own configuration so the next sessions start cheaper.
It treats the agent as a black box
The plugin does not fork the agent and does not patch the model. Every change it makes goes through surfaces the agent already reads:
- Facts are injected as context at session start
- Tools are removed through the agent's own settings files
- A stuck turn gets a redirect note attached to a tool result
Cost is counted the way you are billed, with cache pricing, and the plugin's own model calls are included in its reports. Nothing it learns leaves your machine: state lives in ~/.tokeneconomizer.
Three mechanisms, two on by default
Micro-skills (off by default). Sessions keep rediscovering the same facts: a build step that must run first, python -m pytest because a bare pytest is missing, an environment variable some command needs. The plugin learns such facts from failures that repeat across sessions, plus a proposer model that reviews recent sessions every third session. A fact gets stated at session start once three sessions have confirmed it, and dropped when later sessions contradict it.
Tool pruning (on). Every loaded tool's definition rides along with every request, so an unused tool is a flat per-request tax. A tool that goes unused over 10 prompts spread across at least 2 sessions, or over 5 sessions of any length, gets left out of later sessions through the agent's settings. One command brings it back.
The loop guard (on). When a turn keeps hitting the same failures, a judge model checks whether it is stuck. At 0.7 confidence the agent gets a redirect, at most once per turn. If the turn is still stuck after that, the plugin tells you how to stop it. It never stops a turn itself. There is also a hard backstop: a churning turn that runs past 2.5 times its usual length gets redirected without waiting for the judge.
The part that got my attention: it writes to your settings files
This is the detail config people should know. The tool-pruning deny list is not plugin-internal state. It is written into the repository's own config files:
.claude/settings.local.json (tools, Claude Code)
.codex/config.toml (capability packages, Codex)
.github/copilot/settings.local.json (MCP servers, Copilot CLI)
The write happens only after the repo's sessions have been idle longer than a prompt cache lives, so a resumed conversation does not get its cached prefix rewritten. It takes effect from the next session and is announced at the next session start. toolprune restore gives a tool back for good.
Practically: your settings files now have a second author. If you review config diffs, and you should, expect machine-written deny rules to show up in them. Tools a person relies on (asking you questions, plan mode, subagents, the web) are protected from pruning, and a deny rule written for a tool that has since become protected is rolled back at the next fold.
The numbers, and the one caveat
The best cell was Claude Code with Opus 4.8: 30.9 percent cheaper, rising to 33.2 percent combined with RTK. That same combination carries the one caveat, Pass@1 declined by 1.0 point there (independent coverage). For comparison, the paper reports RTK and RepoMap increased cost in 12 and 15 cells respectively, where TokenEconomizer was cheaper in all 30.
Savings are back-loaded: most repositories cost more at first, and all were cheaper by the sixth task. On Web-Bench it saved 16 to 21 percent, and on two security-auditing benchmarks where every task sits in a new repository, 6.4 to 19.6 percent. The discount scales with repetition. Same repo, recurring work, bigger saving.
Check the version window before you install
The supported-versions table is narrow:
- Claude Code 2.1.300 to 2.1.310
- Codex 0.153.1 to 0.164.0
- Copilot CLI 1.0.83 to 1.0.102
- Pi 1.2.0 to 1.3.0, opencode support experimental
Claude Code shipped 2.1.316 on October 31, so if you auto-update you are already outside the tested window. The plugin needs no API key of its own and never handles credentials. Its judge and proposer calls run through your existing agent login, on your plan, and are counted in its ledger.
What this means for your hand-written rules
Micro-skills are not a machine-written CLAUDE.md. The facts are session-start context, derived per repository, reversible, and never checked in. Your instruction file stays the contract you and your team actually review.
Two consequences are worth thinking through. First, learned facts are exactly the content that bloats rules files: build commands, command spellings, environment variables. If a plugin absorbs that layer, your file can stick to standards and constraints, which is what it is good at. Second, the paper's savings come from repositories where sessions kept rediscovering the same facts. If you already run a lean, accurate rules file, there is less waste left to remove and your savings will land below the headline. Cheap sessions and disciplined config overlap, and that is fine.
Trying it this week
git clone https://github.com/microsoft/TokenEconomizer.git
claude plugin marketplace add "$PWD/TokenEconomizer"
claude plugin install tokeneconomizer@tokeneconomizer
Start claude in a repository you touch daily, run /tokeneconomizer:status, and let a few sessions finish. Once per machine, and again after Claude Code updates, run /tokeneconomizer:toolprune refresh-catalog. The everyday commands:
/tokeneconomizer:status # what it learned here and what sessions cost
/tokeneconomizer:microskills # the facts it injects, and candidates
/tokeneconomizer:toolprune # tools it left out, with the evidence
/tokeneconomizer:config # every setting in effect
Inside the agent a command costs a model turn. From a terminal, ~/.tokeneconomizer/bin/te status in the repository is free. If you keep strict deny rules of your own, diff your settings files after the first few folds. Cost control and cache behavior are the same seam: we wrote about splitting agent context across cache TTLs here, and a plugin that manages what gets re-sent is working the same problem from the tool side.
1 reaction
Microsoft выпустила TokenEconomizer — полезный плагин для экономии
токенов в Claude Code, Codex, Copilot CLI и Pi. Судя по бенчмаркам, он снижает среднюю стоимость работы ИИ-агентов на 18,4-30,9% без потери в качестве. Тулза учится на истории...
microsoft/TokenEconomizer
TokenEconomizer: Experience-Driven Black-Box Cost Control for Coding Agents
microsoft/TokenEconomizer
#ai #tokens #agents #msresearch #oss #softwaredev #mcp #microsoft
1 like
TokenEconomizer (Microsoft Research + UC Berkeley) plugs into Claude Code, Codex, Copilot CLI and Pi and learns from past sessions in a repo: it remembers commands that work there, drops unused tools and nudges the agent out of failure loops. 486 stars.
microsoft/TokenEconomizer
27 likes · 1 repost · 1,472 views
Mentions
Microsoft and UC Berkeley just open-sourced TokenEconomizer (microsoft/TokenEconomizer) to fix token bloat in Claude Code and OpenAI Codex. Tested across 277 SWE-Bench repositories, it cuts total coding agent cost by 18.4% to 30.9% using repo micro-skills memory, dynamic MCP tool pruning, and a dead-end loop guard.
coding agents waste tokens relearning the same repo every session. which test command works, which tools never get used. microsoft + berkeley open-sourced a claude code / codex plugin that learns from past sessions. 18-31% cheaper on swe-bench. microsoft/TokenEconomizer t.co
microsoft open sourced tokeneconomizer, a plugin that makes claude code, codex and copilot cli cheaper 💸 remembers repo facts, drops unused tools, breaks failure loops they claim 18-31% lower cost on swe-bench. agents need memory, not bigger models
microsoft open sourced tokeneconomizer, a plugin that makes claude code, codex and copilot cli cheaper 💸 remembers repo facts, drops unused tools, breaks failure loops they claim 18-31% lower cost on swe-bench. agents need memory, not bigger models microsoft/TokenEconomizer
Microsoft ➡️ "TokenEconomizer: Experience-Driven Black-Box Cost Control for Coding Agents. Our measurement on SWE-Bench tasks: 18.4–30.9% lower mean cost across five agent–model combinations, cheaper in all 30 repository × combination cells, and Pass@1 up in 4 of 5." ➡️ Good for budget? :P microsoft/TokenEconomizer Community Overview: lifehubber.com Resources are shared for discovery and are not independently vetted—please do your own due diligence. New resources are added regularly — feel free to join the sub for updates. Full searchable archive of all resources posted so far on our community site, LifeHubber: lifehubber.com 300+ open-ish AI models, agents, tools, datasets, and related resources, with filtering and sorting.
Microsoft ➡️ "TokenEconomizer: Experience-Driven Black-Box Cost Control for Coding Agents. Our measurement on SWE-Bench tasks: 18.4–30.9% lower mean cost across five agent–model combinations, cheaper in all 30 repository × combination cells, and Pass@1 up in 4 of 5." ➡️ Good for budget? :P microsoft/TokenEconomizer Community Overview: lifehubber.com Resources are shared for discovery and are not independently vetted—please do your own due diligence. New resources are added regularly — feel free to join the sub for updates. Full searchable archive of all resources posted so far on our community site, LifeHubber: lifehubber.com 300+ open-ish AI models, agents, tools, datasets, and related resources, with filtering and sorting.
Microsoft's TokenEconomizer cuts coding agent costs up to 30.9%, and it does it by editing your config files The short version Microsoft Research, UC Berkeley and Microsoft Security published TokenEconomizer last week (v0.1.0, October 26, Apache 2.0). It is a plugin for Claude Code, Codex, the GitHub Copilot CLI and Pi that cuts what your agent costs to run, without touching model weights or agent code. On 277 SWE-Bench tasks across six repositories it lowered mean cost per task by 18.4 to 30.9 percent across five agent-model combinations, it came out cheaper in every one of the 30 repository-by-combination cells, and Pass@1 went up in four of the five combinations (paper). I spend my days on agent config files, so the interesting part for me is not the percentage. It is how the thing works: it learns from your finished sessions, then edits your agent's own configuration so the next sessions start cheaper. It treats the agent as a black box The plugin does not fork the agent and does not patch the model. Every change it makes goes through surfaces the agent already reads: - Facts are injected as context at session start - Tools are removed through the agent's own settings files - A stuck turn gets a redirect note attached to a tool result Cost is counted the way you are billed, with cache pricing, and the plugin's own model calls are included in its reports. Nothing it learns leaves your machine: state lives in ~/.tokeneconomizer. Three mechanisms, two on by default Micro-skills (off by default). Sessions keep rediscovering the same facts: a build step that must run first, python -m pytest because a bare pytest is missing, an environment variable some command needs. The plugin learns such facts from failures that repeat across sessions, plus a proposer model that reviews recent sessions every third session. A fact gets stated at session start once three sessions have confirmed it, and dropped when later sessions contradict it. Tool pruning (on). Every loaded tool's definition rides along with every request, so an unused tool is a flat per-request tax. A tool that goes unused over 10 prompts spread across at least 2 sessions, or over 5 sessions of any length, gets left out of later sessions through the agent's settings. One command brings it back. The loop guard (on). When a turn keeps hitting the same failures, a judge model checks whether it is stuck. At 0.7 confidence the agent gets a redirect, at most once per turn. If the turn is still stuck after that, the plugin tells you how to stop it. It never stops a turn itself. There is also a hard backstop: a churning turn that runs past 2.5 times its usual length gets redirected without waiting for the judge. The part that got my attention: it writes to your settings files This is the detail config people should know. The tool-pruning deny list is not plugin-internal state. It is written into the repository's own config files: .claude/settings.local.json (tools, Claude Code) .codex/config.toml (capability packages, Codex) .github/copilot/settings.local.json (MCP servers, Copilot CLI) The write happens only after the repo's sessions have been idle longer than a prompt cache lives, so a resumed conversation does not get its cached prefix rewritten. It takes effect from the next session and is announced at the next session start. toolprune restore gives a tool back for good. Practically: your settings files now have a second author. If you review config diffs, and you should, expect machine-written deny rules to show up in them. Tools a person relies on (asking you questions, plan mode, subagents, the web) are protected from pruning, and a deny rule written for a tool that has since become protected is rolled back at the next fold. The numbers, and the one caveat The best cell was Claude Code with Opus 4.8: 30.9 percent cheaper, rising to 33.2 percent combined with RTK. That same combination carries the one caveat, Pass@1 declined by 1.0 point there (independent coverage). For comparison, the paper reports RTK and RepoMap increased cost in 12 and 15 cells respectively, where TokenEconomizer was cheaper in all 30. Savings are back-loaded: most repositories cost more at first, and all were cheaper by the sixth task. On Web-Bench it saved 16 to 21 percent, and on two security-auditing benchmarks where every task sits in a new repository, 6.4 to 19.6 percent. The discount scales with repetition. Same repo, recurring work, bigger saving. Check the version window before you install The supported-versions table is narrow: - Claude Code 2.1.300 to 2.1.310 - Codex 0.153.1 to 0.164.0 - Copilot CLI 1.0.83 to 1.0.102 - Pi 1.2.0 to 1.3.0, opencode support experimental Claude Code shipped 2.1.316 on October 31, so if you auto-update you are already outside the tested window. The plugin needs no API key of its own and never handles credentials. Its judge and proposer calls run through your existing agent login, on your plan, and are counted in its ledger. What this means for your hand-written rules Micro-skills are not a machine-written CLAUDE.md. The facts are session-start context, derived per repository, reversible, and never checked in. Your instruction file stays the contract you and your team actually review. Two consequences are worth thinking through. First, learned facts are exactly the content that bloats rules files: build commands, command spellings, environment variables. If a plugin absorbs that layer, your file can stick to standards and constraints, which is what it is good at. Second, the paper's savings come from repositories where sessions kept rediscovering the same facts. If you already run a lean, accurate rules file, there is less waste left to remove and your savings will land below the headline. Cheap sessions and disciplined config overlap, and that is fine. Trying it this week git clone https://github.com/microsoft/TokenEconomizer.git claude plugin marketplace add "$PWD/TokenEconomizer" claude plugin install tokeneconomizer@tokeneconomizer Start claude in a repository you touch daily, run /tokeneconomizer:status, and let a few sessions finish. Once per machine, and again after Claude Code updates, run /tokeneconomizer:toolprune refresh-catalog. The everyday commands: /tokeneconomizer:status # what it learned here and what sessions cost /tokeneconomizer:microskills # the facts it injects, and candidates /tokeneconomizer:toolprune # tools it left out, with the evidence /tokeneconomizer:config # every setting in effect Inside the agent a command costs a model turn. From a terminal, ~/.tokeneconomizer/bin/te status in the repository is free. If you keep strict deny rules of your own, diff your settings files after the first few folds. Cost control and cache behavior are the same seam: we wrote about splitting agent context across cache TTLs here, and a plugin that manages what gets re-sent is working the same problem from the tool side.
Microsoft выпустила TokenEconomizer — полезный плагин для экономии токенов в Claude Code, Codex, Copilot CLI и Pi. Судя по бенчмаркам, он снижает среднюю стоимость работы ИИ-агентов на 18,4-30,9% без потери в качестве. Тулза учится на истории... microsoft/TokenEconomizer
TokenEconomizer: Experience-Driven Black-Box Cost Control for Coding Agents microsoft/TokenEconomizer #ai #tokens #agents #msresearch #oss #softwaredev #mcp #microsoft
TokenEconomizer (Microsoft Research + UC Berkeley) plugs into Claude Code, Codex, Copilot CLI and Pi and learns from past sessions in a repo: it remembers commands that work there, drops unused tools and nudges the agent out of failure loops. 486 stars. microsoft/TokenEconomizer