Snippet Auditor — Development agent for Claude Code
Audits ONE small Python file for correctness bugs and returns a single JSON verdict.
How to install Snippet Auditor
Installs to ~/.claude/agents/asampath1-mock-project-snippet-auditor.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/asampath1/mock-project/HEAD/.claude/agents/snippet-auditor.md -o ~/.claude/agents/asampath1-mock-project-snippet-auditor.md Restart Claude Code, or start a new session, for it to be picked up.
What Snippet Auditor does
name: snippet-auditor description: Audits ONE small Python file for correctness bugs and returns a single JSON verdict. Used by the evaluation labs; read-only. tools: Read
You audit exactly one Python file, given by path in your prompt, for **correctness bugs**: code that returns a wrong result or fails on valid input. Do not report style, naming, formatting, missing docstrings or type hints, or performance.
Read the file, then reply with exactly one line of JSON and nothing else:
{"v
Alternatives in Development
- Review Bugs — Reviews code changes for potential bugs found by single-threaded sequential reasoning: logic errors, null safe 432 ★
- Dashclaw Drift Auditor — Audits DashClaw's drift-prone hardcoded counts and version stamps — SDK method counts (Node/Python), MCP tool/ 297 ★
- Bibliography Auditor — Audits bibliography entries for completeness, consistency, and hygiene — checks .bib files, compiled PDF for u 169 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Pin Auditor
Audits validator checks for VACUITY — a pin that reports green against the mutation it was written to catch. R
Bitness Auditor
Audits one Delphi unit (or a small group) for places that conflate the DEBUGGER's pointer width with the TARGE
Gregorian Decision
Specialist that runs a decision through the full With Love, Math framework and returns a structured verdict. I
Bully Evaluator
Evaluates a single bully semantic-evaluation payload against a diff and returns a structured violation list. I
Memory Codification Auditor
Audits user-scope agent memory for durable conventions that live only in memory and were never codified into t
Recurse Auditor
Audits RecurseKit run artifacts for correctness, completeness, and efficiency.
Related Skills
Claim Audit
A Claude skill that audits AI-written text claim by claim: quotes each checkable claim, tags it with one of te
Five Bugs
A 10-minute smoke test for AI coding agents. Five seeded Python bugs, one checker the agent sees and one it ne
Surrealql Review
Review a SurrealQL file or snippet for correctness, v3 compatibility, and common gotchas.