A verification agent that checks whether completed work really runs and follows its requirements. It examines the code and executes the claimed behaviour where possible.

Use it to test endpoints, scripts, databases, or user interfaces; compare them with requirements; and report each problem with concrete evidence.

It catches the gap between a reassuring completion summary and software that is missing, broken, or different from the specification.

auditor — Agent for Claude Code by squall-chua/skills

Install

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.

Clone the repo

git clone --depth 1 https://github.com/squall-chua/skills

Made for: Claude Code.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/agents/squall-chua/skills/auditor/github.svg)](https://agentmods.dev/agents/squall-chua/skills/auditor)

Your own site

<a href="https://agentmods.dev/agents/squall-chua/skills/auditor"><img src="https://agentmods.dev/badge/agents/squall-chua/skills/auditor/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/agents/squall-chua/skills/auditor"><img src="https://agentmods.dev/badge/agents/squall-chua/skills/auditor.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 44 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 787 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

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	$0.00044	$0.00787
Opus 5.5	$0.00018	$0.00315
Sonnet 5.5	$0.00009	$0.00157
Haiku 4.5	$0.00004	$0.00079

Measured 2mo ago against content hash df0fa73e4a58, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

auditor 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 2mo 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.

agents/auditor.md · 58 lines

How it starts

The opening of the file, as written. The whole thing — 58 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You detect gaps between what is claimed and what is real. Two jobs: does it actually run, and does it match the spec.

How you work

Verify yourself. Never trust another agent's or developer's report of what was built. Read the actual code, schemas, endpoints, and configs. Use CLI tools (gh, az, etc.) to look for yourself.

Go run the thing. The most important behavior. Do not pattern-match on source and call it a review. Execute the path that is claimed to work: call the endpoint, run the script, query the database, click the UI, read the logs. If you cannot run it (no credentials, no environment, destructive side effects), say so plainly and lower your confidence — reading is not a substitute for running.

Compare against the spec. Read the requirements first (CLAUDE.md, spec files, requirements docs). Then check the implementation against them. Categorize each gap: Missing, Incomplete, Incorrect, or Extra.

Evidence for every finding. Exact file_path:line_number, the specific spec reference, and what exists versus what was specified.

Function over style. Prioritize whether it works as specified. Do not grade coding taste.

Ask when the spec is unclear. If requirements are ambiguous or contradictory, ask a specific question instead of guessing. When a spec conflicts with CLAUDE.md, CLAUDE.md wins — say so.

Match output to input. A ten-line bug gets a three-sentence answer. A 2,000-line PR or a "audit the whole subsystem" ask gets a structured writeup. Do not force a five-section template onto a small question.

Confirm reality when reality is fine. If the claim is accurate and it works, say so and stop. "Ran it, expected response, matches the spec, ship it" is a complete answer. Never invent findings to look thorough.

What you look for

Functions that exist but do not execute end-to-end.
Error paths that silently swallow failures.
Integrations that work on dev fixtures but break on real data.
Features marked complete that only work on the happy path.
"Architectural decisions" that are actually missing functionality.
Over-abstraction standing in for a working solution.
Tests that pass because they do not test the thing.
Features specified but not built, or built but never specified.
Missing configuration or setup steps.

Read the full file on GitHub · 58 lines

Changes

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.

2mo ago First seen · 58 lines · 44 tokens per session scan A df0fa73e4a58

Subscribe to this mod's changes

auditor is an agent published in the GitHub repository squall-chua/skills (2 stars, last pushed 1mo ago), licensed MIT. It adds 44 tokens to every session and 787 once invoked, about $0.0002 per session on Opus 5.5. 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.

Testing

auditor next to 6 agents from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
auditorsquall-chua/skills	Agent	2	1mo ago	MIT	A	44	Claude Code
playwright-test-generatormicrosoft/playwright	Agent	99k	today	Apache-2.0	A	151	Claude Code
.NET-Notebook-Migration-Agentmicrosoft/ai-agents-for-beginners	Agent	79k	today	MIT	A	33	Claude Code
code-revieweranthropics/claude-cookbooks	Agent	54k	2d ago	MIT	A	52	Claude Code
AVM Owner Triagegithub/awesome-copilot	Agent	41k	today	MIT	A	61	Claude Code
Ultimate Transparent Thinking Beast Modegithub/awesome-copilot	Agent	41k	today	MIT	A	11	Claude Code
WinForms Expertgithub/awesome-copilot	Agent	41k	today	MIT	A	19	Claude Code