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/akirtok/preflight-security-audit/auditgit clone --depth 1 https://github.com/akirtok/preflight-security-auditWrote 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/akirtok/preflight-security-audit/audit)<a href="https://agentmods.dev/commands/akirtok/preflight-security-audit/audit"><img src="https://agentmods.dev/badge/commands/akirtok/preflight-security-audit/audit.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 | $0.00013 | $0.00674 |
| Opus 5 | $0.00006 | $0.00337 |
| Sonnet 5 | $0.00003 | $0.00135 |
| Haiku 4.5 | $0.00001 | $0.00067 |
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 5d 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.
What it actually says
Run a comprehensive Preflight Security Audit on $1 (default: the current
project root). Follow the methodology in the preflight-security-audit skill —
read ${CLAUDE_PLUGIN_ROOT}/skills/preflight-security-audit/SKILL.md and the
reference files it points to.
Execute these steps:
-
Scope & detect stack. Read
package.json/lockfiles/config and skim the tree to identify languages, frameworks, and services (Next.js, Supabase, Stripe, AI SDKs, email providers, etc.). State the scope and detected stack up front. From the stack, decide which stack-conditional passes apply — serverless/edge (Vercel/Netlify/Lambda/Workers), GraphQL/WebSocket/SSE, Docker (Dockerfile/compose), and JWTs — and skip the ones whose markers are absent (mark them N/A). -
Run all six tracks. Always run every track; stack-conditional passes run only when their marker was detected in step 1 — skip and mark N/A otherwise (keeps scans fast).
- Prefer to dispatch the five auditor agents in parallel with the Task tool:
code-core-auditor,web-security-auditor,ai-security-auditor,privacy-compliance-auditor,launch-readiness-auditor. Give each the scope path and tell it to read its matching reference file. - If subagents are unavailable, run the passes inline by reading each reference
file (
01–05) and sweeping the code yourself.
- Prefer to dispatch the five auditor agents in parallel with the Task tool:
-
Collect every raw finding with:
file:line, track, category, description, exploit/failure path, suggested fix, proposed severity. -
Verify (Track 6). Run the combined findings through the
finding-verifieragent (orreferences/06-verification.mdinline). Reject anything unprovable, merge duplicates, and recalibrate severity per the rubric. -
Write the report to
AUDIT-<YYYY-MM-DD>.mdin the project root using the template in SKILL.md. Include the summary counts, the 0–100 security score (perreferences/06-verification.md), and the ship recommendation (BLOCK / FIX-FIRST / GO-WITH-FOLLOWUPS / GO). -
Report back in chat: the counts, the security score, the ship recommendation, and the top 3 findings. For every Critical and High finding, show the proposed fix as a concrete diff and ask which the user wants applied. Do not edit any code in this command — applying fixes is done by
/audit-fixafter approval.
Be evidence-driven: never report a finding you cannot tie to a concrete line and a real exploit or failure path. A false alarm is as harmful as a miss.
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.
- 5d ago First seen · 50 lines · 13 tokens per session scan A e643ba0ff3bb
audit is a command published in the GitHub repository akirtok/preflight-security-audit (4 stars, last pushed 1mo ago), licensed MIT. It adds 13 tokens to every session and 674 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 2026-08-31.
Other commands, from other repositories
edit-meta-creative
Edit a Meta ad's creative content — landing page URL, primary text, headline, description, CTA, link description — across one ad or many. Use when the user says "change the URL on my ads", "update the copy", "swap the CTA", or anything that changes what an existing Meta ad shows. Meta does not allow in-place creative…
diagnose-underperformer
Diagnose why a specific Google Ads or Meta Ads campaign / ad set / ad is underperforming and prescribe the next concrete action. Use when the user says something like "why is X not converting?", "this campaign is wasting money", or "why are impressions so low?". Read-only — no changes are made.
launch-meta-campaign
Launch a complete Meta Ads campaign — campaign, ad set, and ad(s) — built paused and ready to review. Use when the user wants a new Facebook/Instagram campaign.
launch-pmax
Launch a Google Ads Performance Max campaign with a complete asset group, audience signals, and (for ecom) listing groups. Built paused for review. Use when the user wants a new PMax campaign.
negative-keyword-sweep
Find wasted spend in Google Ads search terms and add negatives at the right scope (account list, campaign, or ad group). Use when the user says "find junk keywords", "stop wasting on bad searches", or wants a search-terms cleanup.
save
Save this conversation as a new or existing reusable context.