claude-token-optimizer
Cut the tokens Claude Code loads at session start. CLI that sets up a small CLAUDE.md with on-demand docs, measures startup tokens, and keeps them small.
Runssource: GitHubhomepageJavaScriptMITcommit a1e437a8c299
JavaScript, MIT licensed. The project labels itself: agents md, ai assistant, automation, claude, claude code, claude md, cli and context engineering.
claude-token-optimizer runs. An Argusic agent installed it in 6 minutes and hit 3 errors and fixed every one without help on a clean machine with no GPU, and the whole session was recorded.
All 414 tests pass, the CLI builds and runs all 10 commands correctly, 5 hook scripts fixed to use bash arithmetic instead of bc, Node upgraded from 18 to 20, and the package installs with only engine warnings
What the agent ended up with on a clean machine, in its own words. How this is measured
Measured by Argusic on a fresh machine every time. Every number links to its evidence. Argusic Score 100 of 100 (the mean of this project's run scores).
At a glance
Subject data from GitHub, linked at the top of this page, refreshed . Test data by Argusic (CC BY 4.0); every number links to a run page with the full log, the recording, and their sha256 hashes.
What was measured
From the run that decided this verdict. Every line is a recorded value, not an estimate.
What went wrong, in the agent's own words
The agent's own words, worst first, never rewritten into advice.
- bc not available in container; 5 hook scripts silently fell back to 0/? token estimates, causing 1 test failure3 minutes
- Node 18 installed but project requires Node 20+2 minutes
- [email protected] and [email protected] require Node >=22, only Node 20 available; EBADENGINE warnings but packages work
- Install time
- 6 minutes
- Cold machine to finish
- 8 minutes
- Errors hit and fixed
- 3 hit, 3 fixed with no human help
- How the result was proved
- npm test exited 0, 414 passed (was 413 before fix); node bin/cto.js --version returned 2.3.20; cto measure, init --yes --framework express, audit, compress, prune, diff, companions all produced correct output
- Model tokens used
- 175,681
- Exact commit tested
- a1e437a8c299
- Container image
- sha256:cdd920bce783
What was not measured
The other half, so you can see exactly how far this evidence goes. A gap is printed as a gap and never filled in.
- Peak GPU memory
- this ran on a CPU machine, so there was no GPU to sample
- Behaviour over a long run
- each attempt uses the smallest documented settings and a fixed time budget, so this measures whether it runs, not how it holds up under load
The end of the recording
The last lines of the terminal session that decided this verdict, exactly as they were recorded.
All verified. Here's the JSON report: **What was done:** - Upgraded Node from v18 → v20.20.2. The project requires Node 20+; it was shipping with v18. - Fixed 5 bash hook scripts (`pre-tool-token-guard.sh`, `session-end-token-report.sh`, `stop-sessio… - After the fix, all 414 tests pass, all CLI commands (`measure`, `init`, `audit`, `compress`, `pru… tokens used 175,681 All verified. Here's the JSON report: **What was done:** - Upgraded Node from v18 → v20.20.2. The project requires Node 20+; it was shipping with v18. - Fixed 5 bash hook scripts (`pre-tool-token-guard.sh`, `session-end-token-report.sh`, `stop-sessio… - After the fix, all 414 tests pass, all CLI commands (`measure`, `init`, `audit`, `compress`, `pru…
Replay the whole session, every command from a clean machine to this point.
How it was tested
One agent, one container, no human help. It installs the project the way its own documentation says to, runs it, and fixes what breaks. Everything below is recorded as it happened: the terminal session, the log and the exact commit. The full method.
Strengths and limits
Measured facts, not opinions. How this is written.
What went well
- Reached a running state on a clean machine, with the session recorded.
- Installed in 6 minutes.
- Recovered from all 3 errors without a human stepping in, which says the failures are documented well enough to solve.
- Ran without a GPU, so it does not need one to start.
- Licensed MIT, as reported by its host.
What did not, or is not known
- Measured once so far. A single attempt shows it can run, not that it runs reliably.
What it is a reasonable choice for
- Trying it on a laptop or a small server: it reached a running state without a GPU.
- A quick evaluation: it was running 6 minutes after the clone.
Also tested, in the same area
Every one of these was installed and run by Argusic on a clean machine. Nothing appears here that was not tested.
Run history
Topics (from GitHub)
agents-mdai-assistantautomationclaudeclaude-codeclaude-mdclicontext-engineeringdeveloper-toolsdocumentationllmnpm-packageopensourcesetup-templatetoken-optimization
Embed the badge
Markdown for the project README. It links back here; terms on the terms page.
[](https://argusic.com/subject/claude-token-optimizer)Questions
- Does claude-token-optimizer run?
- Yes. claude-token-optimizer runs. Argusic installed and launched it on a clean machine in 6 minutes, hitting 3 errors on the way, and recorded the session.
- How did Argusic test claude-token-optimizer?
- On a fresh, disposable machine, with every command recorded and the repository pinned at commit a1e437a8c299. One attempt is recorded, and the full method is on the methodology page.
- What did it cost to test claude-token-optimizer?
- The run that produced this verdict cost $0.02: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.
- How long does claude-token-optimizer take to install?
- 6 minutes in the run that decided this verdict, from a clean machine to a running state. That is one measured attempt, not an average.
- Does claude-token-optimizer need a GPU?
- Not to start. Argusic reached a running state on a machine with no GPU.
- What goes wrong when installing claude-token-optimizer?
- 3 things broke in the recorded run, and 3 were fixed without human help. Each one, and the time it cost, is listed on this page.
- Where is the evidence for claude-token-optimizer?
- The recorded run is on this page, each linking to its full log and terminal recording, stored with a sha256 fingerprint so it cannot be quietly altered.