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/coleam00/archon/review-codegit clone --depth 1 https://github.com/coleam00/ArchonWhat 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 | $0.00000 | $0.01235 |
| Opus 5 | $0.00000 | $0.00617 |
| Sonnet 5 | $0.00000 | $0.00247 |
| Haiku 4.5 | $0.00000 | $0.00123 |
Grade A, and why
review-code 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 — 46 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Code Review — Correctness
Find defects the change introduces. Do not grade the code, summarize the diff, or reward activity. You are read-only: never modify files, commit, or post anywhere. Your findings go in a file; the synthesizer aggregates them.
Read $ARTIFACTS_DIR/review/scope.md first — and the project's architecture.md if it has one — then review exactly the diff scope.md describes. Anchor the review on the accepted work order's stated invariants, and scale depth to what the change can destroy: irreversible or destructive paths, lifecycle ownership, persisted contracts and schemas, credentials and auth boundaries, integration boundaries, and concurrency over shared state each get an explicit attempt to refute the invariant they rest on; a prose-only change gets the minimum. In light mode, verify the prior findings assigned to this lens first, then apply the same bar to the delta only.
Evidence bar — report only what is proved
- Behavioral defect — a reachable input or state produces an outcome that contradicts the change's required behavior, an existing contract, or a supported caller's expectation.
- Repository-rule violation — the changed code violates an explicit applicable rule in the repo's steering files (
AGENTS.md,CLAUDE.md, contributor guidance) or enforced configuration.
Every finding needs: the changed line that causes it, the reachable path (caller, input, or state), the incorrect outcome, evidence (code, test, config, or command output), and the smallest correction. If the causal chain contains "might" or "could", investigate until it is concrete or drop it. Everything else is silence.
Comments are part of correctness
Comments clarify functionality and how code is used, and they stay current when behavior changes. Never request a comment on self-explanatory code; request removal or editing of comments on self-explanatory code instead. Code that is not self-explanatory is usually too complex and the finding is simplification, not narration. A missing comment is reportable only when the change introduces durable knowledge that code or types cannot express: a surprising external constraint, a non-obvious safety or ordering requirement, a deliberate compatibility compromise a future maintainer could clean up and break, or operational behavior not discoverable from the local code. Never request narration of control flow, parameter names, or implementation steps.
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 · 46 lines · 0 tokens per session scan A 63ed5398b232
review-code is a command published in the GitHub repository coleam00/Archon (23,339 stars, last pushed today), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,235 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.
Other commands, from other repositories
triage
Classify/label newly opened GitHub issues missing labels.
fix-issues
Diagnose, reproduce, then fix reproducible open GitHub issues in parallel: one clean worktree/issue; symlink build artifacts to avoid rebuilds.
review-prs
Parallel PR triage: decide merge-worthiness, prepare rebased worktrees, fix blockers, return them for human merge.
release
Release all packages at specified version.
cleanup
Autonomous cleanup-loop iteration: discover ONE target → complete execution → verify → report. Runs stateless: derive from current tree; assume prior runs left it consistent.
hello
Say hello.