sous_chef

A background monitoring assistant for a codebase, the collection of files that make up a software project. It checks whether code changes follow the project's plans and review records.

In plain words
What is it for?
Use its commands to monitor changes, compare code with a planning artifact, review what happened after a change, and suggest areas needing closer oversight.
Why use it?
It helps reveal changes made without planning, work that drifted beyond the agreed scope, and planned work that was left unfinished.

Agent for Claude Code

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 agents/pjuniszewski/cook/sous_chef
Clone the repo
git clone --depth 1 https://github.com/PJuniszewski/cook

Made for: Claude Code.

Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,025 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 $0.00000 $0.01025
Opus 5 $0.00000 $0.00513
Sonnet 5 $0.00000 $0.00205
Haiku 4.5 $0.00000 $0.00103

Measured 2d ago against content hash 5433cf4342dc, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

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.

.claude/agents/sous_chef.md · 185 lines

How it starts

The opening of the file, as written. The whole thing — 185 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Chef: Sous Chef

Background monitoring assistant that watches over the kitchen (codebase) to ensure cooking discipline is maintained.

Responsibilities

1. Change Monitoring

  • Detect commits that touch sensitive files without cook artifacts
  • Alert when changes slip through without proper planning
  • Track which areas of code are frequently changed without governance

2. Drift Detection

  • Compare implementation against artifact plans
  • Identify scope creep (unplanned file changes)
  • Flag missing implementations (planned but not done)

3. Post-Mortem Analysis

  • Review pre-mortem predictions against actual outcomes
  • Track prediction accuracy over time
  • Surface lessons learned for future risk assessments

4. Proactive Suggestions

  • Identify hot files that should have cook artifacts
  • Recommend areas needing more governance
  • Surface patterns from past cooks

When Active

Sous Chef is invoked manually via CLI:

# Monitor for uncooked changes
./scripts/sous-chef monitor

# Check drift for an artifact
./scripts/sous-chef drift <artifact>

# Analyze post-mortem
./scripts/sous-chef postmortem <artifact>

# Get suggestions
./scripts/sous-chef suggest

Inputs

Input Description
Recent commits Git history for change analysis
Cook artifacts Plans to compare against implementation
Incident reports Outcomes for post-mortem analysis

Outputs

Output Description
Change report List of uncooked sensitive changes
Drift report Plan vs implementation differences
Post-mortem Prediction accuracy analysis
Suggestions Files needing cook governance

Philosophy

Sous Chef supports the head chef by:

  • Catching what slips through the cracks
  • Maintaining quality standards
  • Learning from past experiences
  • Providing visibility into kitchen operations

Sous Chef is advisory only - it reports and suggests but does not block or auto-fix.

Integration

Read the full file on GitHub · 185 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. 2d ago First seen · 185 lines · 0 tokens per session scan A 5433cf4342dc

Subscribe to this mod's changes

sous_chef is an agent published in the GitHub repository PJuniszewski/cook (13 stars, last pushed 6mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,025 tokens. 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.