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/kao273183/mk-qa-master/api-securitygit clone --depth 1 https://github.com/kao273183/mk-qa-masterWhat 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 | $0.00031 | $0.01069 |
| Opus 5 | $0.00015 | $0.00535 |
| Sonnet 5 | $0.00006 | $0.00214 |
| Haiku 4.5 | $0.00003 | $0.00107 |
Grade A, and why
api-security 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 — 94 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are operating as the mk-qa-master agent in OWASP API security
scanning mode. Follow Flow 5 in the parent SKILL.md and read
reference/api-security-deep.md before proceeding.
Spec target:
$ARGUMENTS
Required gates BEFORE you call run_api_security_scan
-
Consent. Check
QA_API_SECURITY_CONSENT=trueis set in the host env. If not, the tool will returnconsent_requiredwith a legal disclaimer — surface that disclaimer verbatim to the user; do NOT re-prompt or paraphrase. -
Authorization. If the spec's
servers[0].url(or the user- providedbase_url) points to anything other thanlocalhost/127.0.0.1, the host MUST be inQA_API_SECURITY_AUTHORIZED_DOMAINS. If not set, the tool returnsunauthorized_domain— surface that verbatim too. -
mass_assignment opt-in. The default
categoriesexcludemass_assignmentbecause it mutates server state (it POSTs probe data). Include it ONLY if:- the spec's base_url is localhost (a known fixture, not a real API), OR
- the user explicitly opted in by saying "include mass assignment" / "scan for over-posting" / "test API3".
Steps
-
Survey the spec. Use the MCP tool's
severity_threshold="info"first pass to see ALL endpoints + which rules apply. Don't expose info-level findings to the user — they're noisy. Just check theops_scannedcount looks reasonable (5+ endpoints). -
Decide categories. Default:
["headers", "broken_auth", "bola", "function_authz"]. Add"mass_assignment"only per the opt-in rule above. -
Decide auth. If the user provided a token, pass it as
auth={"token": "..."}. If they provided BOTH user-a + user-b tokens (for BOLA), pass both +bola_test_idsmapping each user to the object ids they own. Withoutbola_test_ids, BOLA emits INFO "skipped" findings rather than firing false positives. -
Run. Call
run_api_security_scan(spec_url=..., auth=..., categories=..., severity_threshold="medium", base_url=..., timeout_s=30).
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 · 94 lines · 31 tokens per session scan A 2f58dcdf23d6
api-security is a command published in the GitHub repository kao273183/mk-qa-master (37 stars, last pushed 3mo ago), licensed MIT. It adds 31 tokens to every session and 1,069 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
flow-nexus-neural
Train and deploy neural networks in distributed sandboxes.
flow-nexus-auth
Flow Nexus authentication and user management.
add-tool
Scaffold a new FreeAgent MCP tool (handler + registration + test) following the repo pattern.
OPSX: Archive
Archive a completed change in the experimental workflow.
session-start
Invoke at the BEGINNING of every working session. Token budget: 500–1,500. -->.
audit-web
请使用 anjian MCP 工具处理这个网站安全评估请求：$ARGUMENTS.