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/telus-labs/stagecraft/auditgit clone --depth 1 https://github.com/telus-labs/stagecraftWhat 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 | $0.00078 | $0.01612 |
| Opus 5 | $0.00039 | $0.00806 |
| Sonnet 5 | $0.00016 | $0.00322 |
| Haiku 4.5 | $0.00008 | $0.00161 |
Grade A, and why
audit 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.
How it starts
The opening of the file, as written. The whole thing — 165 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/audit
You are running a full codebase audit on this project.
Read .claude/skills/audit/SKILL.md before doing anything else. That file defines what each phase does, what each phase produces, and what inputs to gather before starting.
Input
The text after /audit is an optional scope constraint.
/audit— audit the entire codebase./audit src/backend/— focus the audit on one subtree./audit --resume— readdocs/audit/status.jsonand continue from the last completed phase.
Startup
Before any phase work:
- Read
.claude/skills/audit/SKILL.md(the phase definitions — the brain of this command). - Read
CLAUDE.mdif it exists (project-specific instructions). - Read
AGENTS.mdif it exists. - Create
docs/audit/if it doesn't exist. - Check for
docs/audit-extensions.md— note whether it exists; if it does, you'll read it at the start of each phase and append findings under a## Project-Specificheading in the phase's output file. - Run skill step 0.0 — archive any prior completed audit. If
docs/audit/status.jsonexists withcurrent_phase: "phase-3", move the prior audit's files intodocs/audit-archive/<date>-<version>-<context>/and append a row todocs/audit-archive/HISTORY.md. Skip on--resume. Full procedure inskills/audit/SKILL.md§ Phase 0 step 0.0. - Check for
docs/audit/status.json— if resuming, load state. Otherwise write a fresh status file:
{
"started": "<ISO 8601 timestamp>",
"scope": "full" | "scoped to <subtree>",
"phases": {
"phase-0": "pending",
"phase-1": "pending",
"phase-2": "pending",
"phase-3": "pending"
},
"current_phase": "phase-0",
"audited_by": "claude-code"
}
Execution
Run each phase as defined in .claude/skills/audit/SKILL.md. Read each phase's template (.claude/skills/audit/../templates/audit/<NN>-*.md — or copy the structure if your tools don't reach that path) before writing the output file.
Phase 0 — Bootstrap
Run steps 0.0 (archive prior audit), 0.1, 0.2, 0.3 from the skill. Write each output file.
If docs/audit-extensions.md declares Phase 0 extensions, run them and append results under ## Project-Specific in each phase-0 file.
Update status.json: "phase-0": "complete".
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.
- 3d ago First seen · 165 lines · 78 tokens per session scan A 85bcd4af9e2f
audit is a command published in the GitHub repository telus-labs/stagecraft (6 stars, last pushed 6d ago), licensed MIT. It adds 78 tokens to every session and 1,612 once invoked, about $0.0004 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.
Other commands, from other repositories
dev
一键启动 HumHum 开发环境。按顺序执行：.
analyze-repo
Analyze an existing repository's structure, conventions, and guardrails.
dispatcher
Pick the next-best repo to work on across the portfolio — rank free repos, recommend one, claim its lease atomically, and route to the entry command.
test
Run all PandaFilter verification steps in order. Stop and report on first failure. This is the required pre-commit gate.
claude-request
Send a prompt to Claude Code via the LLM gateway with session tracking.
doctor
Health check for existing auth setups. Diagnoses missing env vars, broken routes, dangerous code patterns, middleware issues, OAuth misconfig, and database connectivity.