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.
git clone --depth 1 https://github.com/VorobiovD/airWrote 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/commands/vorobiovd/air/review-self)<a href="https://agentmods.dev/commands/vorobiovd/air/review-self"><img src="https://agentmods.dev/badge/commands/vorobiovd/air/review-self.svg" alt="Measured on agentmods" 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 | $0.00000 | $0.03045 |
| Opus 5 | $0.00000 | $0.01522 |
| Sonnet 5 | $0.00000 | $0.00609 |
| Haiku 4.5 | $0.00000 | $0.00304 |
Grade D, and why
review-self scanned grade D with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 7d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
CODEX_SCRIPT=$(find ~/.claude/plugins/cache/openai-codex -name "codex-companion.mjs" 2>/dev/null | sort -V | tail -1) Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
find /tmp -maxdepth 1 -name 'air-*' -mtime +1 -exec rm -rf {} + 2>/dev/null How it starts
The opening of the file, as written. The whole thing — 226 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Self-Review Flow (--self mode)
When --self is passed, this is a completely different flow. No PR needed. Reviews local changes before push.
Self Step 0: Initialize Session Temp Directory
Before any /tmp write, mint a per-invocation session dir so parallel /air:review --self runs (or a review + self-review in two Claude Code sessions) don't corrupt each other's wiki files and diffs. Claude Code's Bash tool starts a fresh shell per call, so export doesn't persist — capture the literal path from the command below and substitute it into every $AIR_TMP reference downstream.
If the orchestrator already minted $AIR_TMP (e.g. /air:review --self routed through review.md Step 0), reuse it — don't double-mint. Otherwise mint a fresh dir:
if [ -z "$AIR_TMP" ]; then
find /tmp -maxdepth 1 -name 'air-*' -mtime +1 -exec rm -rf {} + 2>/dev/null
AIR_TMP=$(mktemp -d "/tmp/air-self-XXXXXX")
fi
# Plugin root for the meta.py invocations in the wiki-learn block below.
if [ -z "${AIR_PLUGIN_ROOT:-}" ]; then
AIR_PLUGIN_ROOT=$(ls -1d ~/.claude/plugins/cache/air/air/*/ 2>/dev/null | sort -V | tail -1 | sed 's:/$::')
fi
if [ -z "$AIR_PLUGIN_ROOT" ] || [ ! -d "$AIR_PLUGIN_ROOT" ]; then
echo "warning: AIR_PLUGIN_ROOT not resolvable; Self Step 7's auto-trigger counter will not increment this run" >&2
AIR_PLUGIN_ROOT=""
fi
echo "$AIR_TMP"
Self Step 1: Get the diff
git diff HEAD > $AIR_TMP/self-review.diff
If the diff is empty, try staged only:
git diff --cached > $AIR_TMP/self-review.diff
If still empty: "No changes to review. Stage or modify files first." and STOP.
Print summary: " files changed, +/-"
Self Step 2: Load Context
Same as regular Step 3 — clone the wiki (into $AIR_TMP/review-wiki-self) and copy ALL wiki pages to $AIR_TMP/ (REVIEW.md, REVIEW-HISTORY.md, PROJECT-PROFILE.md, ACCEPTED-PATTERNS.md, SEVERITY-CALIBRATION.md, GLOSSARY.md). Also read CLAUDE.md (or AGENTS.md if there is no CLAUDE.md) from the repo root and the current repo's .claude/agents/ for any repo-specific review rules. Run Step 3.5 (first-run project discovery — see commands/review.md Step 3.5) if PROJECT-PROFILE.md doesn't exist.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 7d ago First seen · 226 lines · 0 tokens per session scan D 8ef2a2441a02
review-self is a command published in the GitHub repository VorobiovD/air (5 stars, last pushed 12d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 3,045 tokens. A static security scan graded it D with 2 findings (reads agent configuration directories, recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
devkit.github.review-pr
Provides comprehensive GitHub pull request review with code quality, security, and best practices analysis. Use when reviewing a PR before merging.
git
The pre-finish status: branch, hygiene findings, message checks, workflow lint, template state.
code-review
Code review for branch changes. Analyzes git diff between branches with multi-level depth (low/medium/high). Matches changes against task description. Returns structured report with severity levels and verdict.
merge-conflict-analysis
You are analyzing merge conflicts for PR #${{ pr-number }}.
repo-audit
Audit a codebase (local or remote GitHub/GitLab) against architecture principles and requirements, surfacing drift, risk, and missing decisions.
advanced-code-review-context
Advanced Code Review Phase 2: Context Analysis - load previous reviews, PR history, declined items.