factory-kit-audit
Skill9 starsChanged 4 months agoMITView on GitHub
Measure the factory-kit's token footprint — baseline vs on-demand, heaviest assets, trim candidates
In this repository42 other files it gives its agents.
- guidance/claude.mdCLAUDE.md
- factory-animationSkill
- factory-api-route-engineerSkill
- factory-apiSkill
- factory-authSkill
- factory-auth-wiring-specialistSkill
- factory-ciSkill
- factory-closeSkill
- factory-code-reviewerSkill
- factory-commitsSkill
- factory-data-layerSkill
- factory-data-pipeline-engineerSkill
- factory-data-pipelinesSkill
- factory-db-migration-engineerSkill
- factory-db-migrationSkill
- factory-db-schema-architectSkill
- factory-deploymentSkill
- factory-designSkill
- factory-entrySkill
- factory-feature-architectSkill
- factory-forms-builderSkill
- factory-formsSkill
- factory-frontend-engineerSkill
- factory-frontendSkill
- factory-llm-workflow-engineerSkill
- factory-llm-workflowsSkill
- factory-observabilitySkill
- factory-pitfallsSkill
- factory-promptingSkill
- factory-promptSkill
- factory-releaseSkill
- factory-security-engineerSkill
- factory-securitySkill
- factory-setup-linearSkill
- factory-stackSkill
- factory-standupSkill
- factory-state-modelingSkill
- factory-submitSkill
- factory-testingSkill
- factory-verification-engineerSkill
- factory-verificationSkill
- factory-voiceSkill
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 1 is distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 13
If the kit can't be found, tell the user to run the kit's install.sh first and stop.
- Distinctive: line 24
No further commentary. The numbers speak.
- Very specific: line 28
Follow factory-voice.md. This is a diagnostic — terse, numbers-first, no narration.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
You're auditing the factory-kit's token cost. This is a read-only report — no files change, no symlinks touched.
What to do
Resolve the kit root. The script lives at
bin/kit-audit.shin the installed kit. Prefer the host's current-skill-directory capability and walk up fromskills/factory-kit-audit/to the plugin root. For symlink installs, resolve the currentSKILL.mdsymlink before walking up. Do not assume a fixed home-directory checkout.If the kit can't be found, tell the user to run the kit's
install.shfirst and stop.- Run the audit. Execute
<kit_root>/bin/kit-audit.shthrough the host's shell capability. The script prints the report to stdout. - Show the output verbatim in a fenced code block. Don't reformat it — the column alignment matters.
One-line takeaway. Below the code block, a single sentence calling out what matters most. Pick from:
- Baseline cost framing — "Baseline is Xk tokens per session; everything else is on-demand."
- Outlier flag — if the trim section named outliers, restate the action
- All-good — "Kit is balanced; no trim work needed."
No further commentary. The numbers speak; you point at the one thing worth doing.
Style
Follow factory-voice.md. This is a diagnostic — terse, numbers-first, no narration. If the user asks "why X is so large," then dig in. Don't volunteer interpretation beyond the one-line takeaway.
The script's estimate is ~4 chars per token. Real tokenizer counts differ by 10-20%. Don't overstate precision — the relative shape across assets is what's actionable, not the absolute numbers.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

