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.
npx agentmods add commands/pjuniszewski/cook/sous-chefgit clone --depth 1 https://github.com/PJuniszewski/cookWhat 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 | $0.00017 | $0.00690 |
| Opus 5 | $0.00009 | $0.00345 |
| Sonnet 5 | $0.00003 | $0.00138 |
| Haiku 4.5 | $0.00002 | $0.00069 |
Grade A, and why
sous-chef 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 2d 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 — 120 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/sous-chef Command
Kitchen assistant that monitors cooking discipline and detects issues.
Syntax
/sous-chef Run change monitor (default)
/sous-chef monitor Detect uncooked sensitive changes
/sous-chef drift <artifact> Compare plan vs implementation
/sous-chef postmortem <artifact> Analyze pre-mortem accuracy
/sous-chef suggest Suggest files needing governance
Options
| Option | Description |
|---|---|
--since <ref> |
Git ref to start from (default: HEAD~10) |
--range <range> |
Git range for drift detection |
--incidents <list> |
Known incidents for postmortem |
Execution
Monitor (default)
Run change monitor to find uncooked sensitive changes:
./scripts/sous-chef monitor
After running, analyze the output:
- If uncooked changes found, ask user: "Found X commits with uncooked changes. Want me to
/cookthem?" - If clean, report: "Kitchen is clean - no uncooked sensitive changes."
Drift Detection
When user specifies an artifact:
./scripts/sous-chef drift "cook/<artifact>.cook.md"
After running, analyze:
- If drift detected, list scope creep and missing files
- Offer: "Want me to update the artifact with actual changes?"
Postmortem
Analyze pre-mortem predictions:
./scripts/sous-chef postmortem "cook/<artifact>.cook.md"
After running, summarize:
- Which predictions came true
- Which were false positives
- Suggestions for future risk assessment
Suggest
Find files that need more governance:
./scripts/sous-chef suggest
After running, offer to create cook artifacts for suggested files.
Interactive Flow
After each command, offer relevant follow-up actions:
Monitor found issues:
Found 2 commits with uncooked sensitive changes:
- abc1234: src/auth/login.ts
- def5678: api/payments.ts
Options:
1. /cook the auth changes
2. /cook the payment changes
3. Ignore (add to exceptions)
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.
- 2d ago First seen · 120 lines · 17 tokens per session scan A 2e71b6ddfa6e
sous-chef is a command published in the GitHub repository PJuniszewski/cook (13 stars, last pushed 6mo ago), licensed MIT. It adds 17 tokens to every session and 690 once invoked, about $0.0001 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 2026-08-30.
Other commands, from other repositories
build
Implement tasks incrementally — build, test, verify, commit. Add "auto" to run the whole plan in one approved pass.
constraints
Define and enforce this project's quality bar — interview, sane defaults, CONSTRAINTS.md.
webperf
Run a web performance audit via the web-performance-auditor persona.
code-simplify
Simplify code for clarity and maintainability — reduce complexity without changing behavior.
plan
Break work into small verifiable tasks with acceptance criteria and dependency ordering.
spec
Start spec-driven development — write a structured specification before writing code.