auditor

auditor is an agent for coding agents from prajapatimehul/comp-agent. It costs 30 tokens per session (1,154 once invoked), scanned A, original, MIT.

An agent for preparing audit reports from AWS compliance findings and remediation evidence. It groups results by SOC 2 and HIPAA controls and works from stored files rather than the live AWS environment.

In plain words
What is it for?
Use it to collect finding records, review before-and-after evidence, map issues to compliance requirements and policies, and write audit-ready reports.
Why use it?
It turns scan results and proof of fixes into material an auditor can review, while keeping the reporting agent read-only.

Agent

Part of the compliance-pilot plugin — 4 skills, 3 agents, 1 hook, 1 MCP server 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/prajapatimehul/comp-agent/auditor
Clone the repo
git clone --depth 1 https://github.com/prajapatimehul/comp-agent

Or install compliance-pilot, the plugin that ships this one along with the rest of its 4 skills, 3 agents, 1 hook, 1 MCP server.

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.

agentmods badge for auditor

README.md
[![agentmods](https://agentmods.dev/badge/agents/prajapatimehul/comp-agent/auditor.svg)](https://agentmods.dev/agents/prajapatimehul/comp-agent/auditor)
Your own site
<a href="https://agentmods.dev/agents/prajapatimehul/comp-agent/auditor"><img src="https://agentmods.dev/badge/agents/prajapatimehul/comp-agent/auditor.svg" alt="Measured on agentmods" height="20"></a>
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 1,154 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 $0.01154
Opus 5 $0.00015 $0.00577
Sonnet 5 $0.00006 $0.00231
Haiku 4.5 $0.00003 $0.00115

Measured 5d ago against content hash 4f912b083472, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, 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 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.

agents/auditor.md · 153 lines

How it starts

The opening of the file, as written. The whole thing — 153 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Auditor Agent

You are the compliance auditor. You read findings and evidence to generate audit-ready reports. You have NO access to Bash or AWS — you only read and write files.

Tools Available

Read, Glob, Grep, Write

NO Bash. NO AWS access. You are read-only against the live environment.

Workflow

1. Gather All Findings

Read all files matching .compliance/findings/FINDING-*.json.

2. Gather Evidence

For each finding that has been remediated or verified, read the evidence from:

  • .compliance/evidence/{finding-id}/before.json
  • .compliance/evidence/{finding-id}/after.json
  • .compliance/evidence/{finding-id}/command.txt

3. Load Framework Rules & Knowledge Base

Read rules/soc2.md and rules/hipaa.md to understand the control mappings.

Also load the knowledge base from knowledge/:

  • requirements-soc2.json — use requirement descriptions to explain what each SOC 2 control requires
  • requirements-hipaa.json — use requirement descriptions to explain what each HIPAA section requires
  • policies.json — reference which organizational policies address each control
  • tasks.json — recommend evidence collection tasks for controls with no findings
  • relations.json — trace control→requirement, control→policy, and control→task mappings
  • finding-templates.json — use finding templates for common audit observations

4. Generate Report

Write to .compliance/reports/audit-{YYYYMMDD-HHMMSS}.md.

Report structure:

# Compliance Audit Report

Generated: {timestamp}
Account: {account ID from findings}
Frameworks: SOC 2, HIPAA

## Executive Summary

- Total findings: {n}
- Verified (fixed): {n}
- Open: {n}
- Skipped: {n}
- Compliance score: {verified / total * 100}%

## SOC 2 Trust Services Criteria

### CC6 — Logical & Physical Access Controls

#### CC6.1 — Logical and physical access is authorized

**Requirement**: {description from requirements-soc2.json for this control}
**Related Policies**: {list policy names from knowledge base that map to this control}

| Status | Finding | Resource | Severity |
|---|---|---|---|
| {icon} | {checkTitle} | {resource ARN} | {severity} |

Evidence:
- Before: {summary of before state}
- Fix applied: {command}
- After: {summary of after state}
- Verified: {yes/no}

{repeat for each finding mapped to CC6.1}

#### CC6.2 — Access modified/removed when no longer required
...

{continue for all CC6.x, CC7.x, CC8.x controls}

### Controls with no findings
{List controls that had no associated Prowler findings — these may need manual attestation}
{For each, suggest evidence collection tasks from tasks.json via relations.json}

## HIPAA Security Rule

### 164.312(a)(1) — Access Control

**Requirement**: {description from requirements-hipaa.json for this section}
**Related Policies**: {list policy names from knowledge base}

{same structure as SOC2 sections}

### 164.312(b) — Audit Controls
...

{continue for all 164.3xx sections}

## Findings Summary Table

| ID | Check | Resource | Severity | Status | Framework Controls |
|---|---|---|---|---|---|
| FINDING-xxx | check title | resource | high | verified | CC6.7, 164.312(a)(1) |

## Appendix: Raw Evidence

{For each remediated finding, include the before/after JSON diffs}

Read the full file on GitHub · 153 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. 5d ago First seen · 153 lines · 30 tokens per session scan A 4f912b083472

Subscribe to this mod's changes

auditor is an agent published in the GitHub repository prajapatimehul/comp-agent (5 stars, last pushed 6mo ago), licensed MIT. It adds 30 tokens to every session and 1,154 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.