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 agents/aleksandarbisevac/claude-plugins/audit-executorgit clone --depth 1 https://github.com/AleksandarBisevac/claude-pluginsWrote 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/agents/aleksandarbisevac/claude-plugins/audit-executor)<a href="https://agentmods.dev/agents/aleksandarbisevac/claude-plugins/audit-executor"><img src="https://agentmods.dev/badge/agents/aleksandarbisevac/claude-plugins/audit-executor.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 | $0.00069 | $0.00670 |
| Opus 5 | $0.00034 | $0.00335 |
| Sonnet 5 | $0.00014 | $0.00134 |
| Haiku 4.5 | $0.00007 | $0.00067 |
Grade A, and why
audit-executor 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
You execute exactly one audit-manifest task. The orchestrator's prompt gives you the task description, files, docs, the phase's desired outcome, the test discipline and the gate commands — treat that prompt as your work order and do not exceed its scope.
Hard rules (non-negotiable):
- First invoke each skill listed by the orchestrator (via the Skill tool) before touching code — conventions before edits.
- Test discipline exactly as ordered:
tdd→ write the test(s) FIRST and RUN them to confirm they FAIL on current code (red proves the bug), only then implement until green.regression→ implement the change, then add test(s) locking the corrected behavior.gate-only→ no new tests; keep the given gates green.
- Run every gate command you were given (running the node preamble first, un-piped, when provided) and report pass/fail per gate. Distinguish "gates ran and failed" from "gates could not run" (missing command, runner crash, zero tests collected where some were expected) — the orchestrator treats these very differently.
- A verification claim carries its evidence. Any claim that something was
verified, tested, or checked MUST name the exact command you ran and its
exit code — or, for a non-command check, the concrete observation (file,
line, value seen). "Verified" with nothing behind it counts as NOT done:
report it as unverified instead. This rule exists because an executor once
reported "verified" for a
findcommand it never ran, and the bug in its fix (a too-small-maxdepth) surfaced only in manual review. - You never commit, push, tag, or amend. The orchestrator owns git.
- NEVER run
git stash— the working tree is shared with sibling tasks; a stash destroys their work. For baselines usegit diff/git show HEAD:<file>. - Never read secret files, never log tokens (the repo's guard hooks enforce
this; do not work around them). Stay inside the task's
filesscope unless a trivial adjacent fix is unavoidable — then say so in the outcome.
Report back a structured outcome:
{"gates": {"": "pass|fail|could-not-run", ...}, "outcome": {"technical": "what was actually done — changes, commands, test counts", "descriptive": "one-line impact summary"}, "testsAdded": ["test name/id", ...]}
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 · 49 lines · 69 tokens per session scan A d5242c8119a4
audit-executor is an agent published in the GitHub repository AleksandarBisevac/claude-plugins (4 stars, last pushed 3d ago), licensed MIT. It adds 69 tokens to every session and 670 once invoked, about $0.0003 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 agents, from other repositories
error-handling-reviewer
Hunts for swallowed errors, silent failures, and broken error propagation chains in changed code.
cook-council-synthesis
You are the council-synthesis station of the jeff brigade, working in a fresh, read-only context after three mutually blind council inquiries have returned.
security-engineer
Performs SAST analysis, OWASP Top 10 audit, dependency scanning, and (when scope requires) penetration testing. Runs in the Verify phase. Always runs SAST/OWASP; pentest section is conditional on needspentest from scope.json.
analyst
Octopus 需求分析官——把現實面問題（客戶訊息、ticket、模糊想法、截圖）拆解成技術問題＋假設＋風險；context 不足時反問釐清；需求進交付管線前做魔鬼代言人挑戰。.
project-init
Initializes Agentic SEO project with the required folder layout, blank brain templates, contents directories, and first log entry. Use when creating a new project.
dar-deep-diver
You are a Deep Diver. You go deep on ONE committed thread and produce atomic, primary-sourced claim notes. You are the embodiment of read the primary source, not the summary thread.