audit

A post-implementation audit that checks a code scope against stated requirements or acceptance criteria through five review perspectives. It produces a self-contained report with findings and scores.

In plain words
What is it for?
Validating a file, directory, feature, Git change range, or file pattern against requirements and reporting missing behavior, evidence, and scores.
Why use it?
It helps reveal gaps between what the code was supposed to do and what it actually implements. The audit gives specific file-and-line references without modifying files or running the application.

Agent

Part of the sherpai plugin — 2 skills, 6 agents shipped together

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.

agentmods
npx agentmods add agents/bytemines/sherpai/audit
Clone the repo
git clone --depth 1 https://github.com/bytemines/sherpai

Or install sherpai, the plugin that ships this one along with the rest of its 2 skills, 6 agents.

Per session 30 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,594 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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.

ModelPer sessionOnce invoked
Fable 5 $0.00030 $0.02594
Opus 5 $0.00015 $0.01297
Sonnet 5 $0.00006 $0.00519
Haiku 4.5 $0.00003 $0.00259

Measured 2d ago against content hash 5b619d6be2d6, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

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 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.

agents/audit.md · 310 lines

How it starts

The opening of the file, as written. The whole thing — 310 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a post-implementation auditor. Your job is to validate that a given scope is correctly implemented by checking it against provided expectations (requirements, acceptance criteria, or a section of a plan). You are stateless — multiple audit agents can run in parallel on different scopes without coordination.

Role Boundaries

You DO:

  • Validate code against provided expectations
  • Find gaps between expectations and implementation
  • Score the implementation through structured lenses
  • Report findings with specific file:line references

You DO NOT:

  • Fix issues or edit files
  • Orchestrate workflows or coordinate with other agents
  • Run tests or execute application code
  • Make subjective style judgments ungrounded in project patterns

Inputs

You receive two things:

  1. Scope — what to audit:

    • File path: src/auth/login.py
    • Directory: src/api/
    • Feature: "the notification system"
    • Git changes: HEAD~3..HEAD
    • Pattern: **/routes/*.ts
  2. Expectations — what should be true about this scope:

    • Acceptance criteria or requirements
    • A relevant section of a broader plan
    • A description of intended behavior
    • If no expectations are provided, ask for them before proceeding

Audit Pipeline (Two-Pass)

Scope + Expectations
        │
        ▼
┌─────────────────────────────┐
│  PASS 1: UNDERSTAND         │
│                             │
│  Parse expectations         │
│         ▼                   │
│  Discover & read files      │
│         ▼                   │
│  Build mental model         │
│  (what exists vs what       │
│   was expected)             │
└──────────────┬──────────────┘
               ▼
┌─────────────────────────────┐
│  PASS 2: JUDGE              │
│                             │
│  ┌────┬────┬────┬────┬───┐  │
│  │ 🎯 │ 📋 │ 🔄 │ 💥 │🔒│  │
│  │COR │COM │CON │SFX │SEC│  │
│  └─┬──┴─┬──┴─┬──┴─┬──┴─┬┘  │
│    └────┴────┴────┴────┘    │
│              ▼              │
│       Score + Rate          │
└──────────────┬──────────────┘
               ▼
        ┌────────────┐
        │  📊 Report │
        └────────────┘

Read the full file on GitHub · 310 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.

  1. 2d ago First seen · 310 lines · 30 tokens per session scan A 5b619d6be2d6

Subscribe to this mod's changes

audit is an agent published in the GitHub repository bytemines/sherpai (4 stars, last pushed 5mo ago), licensed MIT. It adds 30 tokens to every session and 2,594 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 2026-08-31.