Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/fledgeling-co/fledgeling-pluginsnpx agentmods add skills/fledgeling-co/fledgeling-plugins/ratchetWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/fledgeling-co/fledgeling-plugins/ratchet)<a href="https://agentmods.dev/skills/fledgeling-co/fledgeling-plugins/ratchet"><img src="https://agentmods.dev/badge/skills/fledgeling-co/fledgeling-plugins/ratchet/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/fledgeling-co/fledgeling-plugins/ratchet"><img src="https://agentmods.dev/badge/skills/fledgeling-co/fledgeling-plugins/ratchet.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00087 | $0.01107 |
| Opus 5.5 | $0.00035 | $0.00443 |
| Sonnet 5.5 | $0.00017 | $0.00221 |
| Haiku 4.5 | $0.00009 | $0.00111 |
Grade A, and why
ratchet scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 95 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Ratchet — the governor
This is what makes the plugin a system rather than a checklist. It reads what the other planes produced, decides what each class has earned, and takes authority away on its own.
The asymmetry is deliberate. Revocations apply immediately and need no signature. Promotions are
written as proposals, because a class gaining authority is a policy change and warrant:charter
owns those.
Running as a Gemini model? Read gemini.md in this directory first, then follow this file with the overrides it names. It says the tier table is the script's output rather than yours, that exit 4 is a completed revocation and not a retry, and that restoring a tier is never one of the available actions. Other models skip it.
Procedure
-
Run it after anything that produces evidence.
python3 scripts/ratchet.py --root <repo>It reads
.warrant/suite-health.json, the regression result,.warrant/oracle-coverage.json,.warrant/escapes.jsonl,.warrant/warrant.tomland.warrant/lanes.toml. -
Read the exit code. Exit 0 means nothing changed or a promotion was proposed. Exit 4 means a revocation fired and has already been applied.
-
Check the control chart on a cadence.
python3 scripts/westgard.py --root <repo> --series <corpus-history.json>A multirule chart over the regression corpus pass rate across runs: 1-3s, 2-2s, R-4s, 4-1s and 10-x. The multirule form is not decoration — a single-threshold alarm on a true-negative-heavy queue either never fires or fires constantly, which is why clinical laboratories stopped using one.
-
Revoke by hand when you have a reason the script cannot see.
python3 scripts/revoke.py --root <repo> --class <name> --reason '<why>'The reason is required. A revocation with no recorded reason is indistinguishable later from a mistake.
The six revocation triggers
| Trigger | Why it revokes |
|---|---|
| a pinned model id or version differs from the one recorded against the class's last regression run | the control changed, so the benchmark no longer holds (C12) |
| the regression run is failing | the machine no longer catches something it has already missed once |
| a new escape in a class above tier 0 | the class was closing items it should not have been |
| a Westgard violation | the corpus pass rate is drifting rather than dipping |
| oracle coverage below the class's tier-1 threshold | the deterministic plane stopped covering the surface |
the class's calibration is older than [staleness] calibration_max_days, or it has none |
evidence that has gone stale is not evidence, and a class whose age cannot be established is treated as stale |
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 95 lines · 87 tokens per session scan A eba51c4b6410
ratchet is a skill published in the GitHub repository fledgeling-co/fledgeling-plugins (2 stars, last pushed 4d ago), licensed MIT. It adds 87 tokens to every session and 1,107 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-04.
Other skills, from other repositories
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…