ratchet

ratchet is a command for coding agents from askwigconsulting/cohort. It costs 33 tokens per session (828 once invoked), scanned A, original, MIT.

A command for repeated, metric-based code optimization in a temporary Git worktree. It proposes a change, runs a fixed evaluator, keeps changes that improve the measured result, and reverts changes that do not.

In plain words
What is it for?
Use it when progress can be measured by a benchmark score, performance number, failing-test count, lint count, or another clearly defined metric.
Why use it?
It automates tedious experiments while keeping the final merge under human review. A fixed evaluator is a command whose measurement the optimization loop cannot alter.

Command

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/askwigconsulting/cohort/ratchet
Clone the repo
git clone --depth 1 https://github.com/askwigconsulting/cohort

Wrote 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.

agentmods badge for ratchet

README.md
[![agentmods](https://agentmods.dev/badge/commands/askwigconsulting/cohort/ratchet.svg)](https://agentmods.dev/commands/askwigconsulting/cohort/ratchet)
Your own site
<a href="https://agentmods.dev/commands/askwigconsulting/cohort/ratchet"><img src="https://agentmods.dev/badge/commands/askwigconsulting/cohort/ratchet.svg" alt="Measured on agentmods" height="20"></a>
Per session 33 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 828 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00033 $0.00828
Opus 5 $0.00016 $0.00414
Sonnet 5 $0.00007 $0.00166
Haiku 4.5 $0.00003 $0.00083

Measured 3d ago against content hash 8a91d42d2735, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

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 3d 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.

canonical/commands/ratchet.md · 69 lines

How it starts

The opening of the file, as written. The whole thing — 69 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Point it at a number and let it climb. /ratchet is a metric-gated optimization loop — Karpathy's AutoResearch pattern (propose a change, run a fixed-budget evaluator, keep the commit if the metric improved, git reset if not, repeat) adapted to Cohort's human gate: the whole climb runs inside a throwaway git worktree, bounded by a budget, and you review the staircase and merge via PR. The autonomy is the inner loop; the merge stays gated. Reach for it when the win is measurable and the search is tedious — a perf number, a benchmark score, a failing-test count, a lint count.

Runs on a coordinator tier (Fable or Opus): you set up the contract and read the staircase; the loop does the methodical climbing.

The three-part contract

Make these three things explicit before you start — it is what makes autonomy safe:

  • The immutable evaluator — one command that prints the objective number, and that the loop never edits. This is the ground truth (Karpathy's prepare.py). If the doer could change the evaluator, it could optimize the metric by lying; it can't, because it only ever touches the worktree's tracked code, which you review.
  • The sandbox — a detached worktree off HEAD. Every proposal lands only here; your working tree is never touched, and a bad run is thrown away.
  • The direction — the objective in words. Keep it tight and surgical ("lower p99 latency in handler.py; change nothing else").

Run it

cohort engine ratchet gpt \
  --evaluator "pytest tests/bench.py -q 2>&1 | tail -1" \
  --metric-regex 'score=([0-9.]+)' \
  --goal maximize --budget 15 --footprint src/handler.py
  • gpt (Codex, edits under its own sandbox) or grok (egress-gated agentic patch) does the proposing; the loop's keep/revert, worktree, ledger, and budget are enforced in code.
  • Each iteration is fed the current best and the recent ledger so it calibrates what to try next — the ratchet-results.tsv staircase is the loop's memory.
  • Ties and non-improvements revert. The lineage only advances on a real gain.

Read the full file on GitHub · 69 lines

Changes

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.

  1. 3d ago First seen · 69 lines · 33 tokens per session scan A 8a91d42d2735

Subscribe to this mod's changes

ratchet is a command published in the GitHub repository askwigconsulting/cohort (2 stars, last pushed 26d ago), licensed MIT. It adds 33 tokens to every session and 828 once invoked, about $0.0002 per session on Opus 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 2026-08-31.