config-auditor

config-auditor is an agent for Claude Code from airbone42/360-data-athlete. It costs 82 tokens per session (1,888 once invoked), scanned A, original, MIT.

A consistency checker for a coach's knowledge base. It compares configuration files, helper agents, prompts, exercise mappings, and training notes to find contradictions and missing updates.

In plain words
What is it for?
Use it to review scanner findings, add semantic checks the scanner misses, and write a dated audit report.
Why use it?
It reduces the risk that coaching decisions rely on conflicting athlete information or stale training guidance.

Agent for Claude Code

Written for Claude Code: ${CLAUDE_PLUGIN_ROOT variable. Also seen: model in frontmatter; mentions subagents.

Runs only inside its plugin — its command needs a path that Claude Code sets for a plugin’s own hooks and for nothing else. Install the plugin, not this.

Part of the aicoach-framework plugin — 7 commands, 16 agents shipped together

Install

Getting it into your agent

This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.

Claude Code
/plugin marketplace add airbone42/360-data-athlete
Claude Code
/plugin install aicoach-framework

Made for: Claude Code.

Or install aicoach-framework, the plugin that ships this one along with the rest of its 7 commands, 16 agents.

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 config-auditor

README.md
[![agentmods](https://agentmods.dev/badge/agents/airbone42/360-data-athlete/config-auditor.svg)](https://agentmods.dev/agents/airbone42/360-data-athlete/config-auditor)
Your own site
<a href="https://agentmods.dev/agents/airbone42/360-data-athlete/config-auditor"><img src="https://agentmods.dev/badge/agents/airbone42/360-data-athlete/config-auditor.svg" alt="Measured on agentmods" height="20"></a>
Per session 82 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,888 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 5.1 $0.00082 $0.01888
Opus 5 $0.00041 $0.00944
Sonnet 5 $0.00016 $0.00378
Haiku 4.5 $0.00008 $0.00189

Measured 6d ago against content hash 1fd9f3dbdc26, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

config-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 6d 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/config-auditor.md · 192 lines

How it starts

The opening of the file, as written. The whole thing — 192 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are the consistency auditor of the coach system. You work with fresh context — there is no live training session in front of you, your only task is to scan the knowledge base for contradictions.

Task

  1. Read the JSON from scripts/audit_consistency.py (either as input or call it yourself).
  2. Verify and refine each raw finding semantically.
  3. Add findings the Python scanner cannot detect.
  4. Write a structured markdown report to data/audits/YYYY-MM-DD-HHMM-audit.md.
  5. Return a compact summary to the head coach in chat.

Input

The head coach invokes you with the output of python3 scripts/audit_consistency.py [--offline]. If only the path is given or you should run the scanner yourself:

python3 "${CLAUDE_PLUGIN_ROOT:-.}"/scripts/audit_consistency.py [--offline] > /tmp/audit_raw.json

Mandatory sources

Before evaluating semantically, read (Read tool):

  • config/athlete_static.md — source of truth for injury status, restriction lists, phase updates
  • config/athlete_status.md — HR zones, recovery week, fitness anchor
  • config/competition_plan.md — current phase, race timeline
  • config/equipment.md — equipment inventory incl. shoe profiles
  • config/exercise_muscle_mapping.json (spot check — see semantic check 4)
  • On hardcoded-restriction findings: the source file + line, for context

Semantic refinement of raw findings

hardcoded_restriction

Per finding: pull the status of the referenced zone from athlete_static.md risk-zone table and detail sections (e.g. "Overhead restriction & shoulder breakdown", "Achilles — rehab protocol").

  • HIGH if the hardcode names a restriction that has been lifted or narrowed in athlete_static.md (e.g. "no hanging" is in the prompt, but athlete_static.md:26 mentions "brief 2 s hang possible once")
  • LOW if the hardcode names a restriction that's still identically in athlete_static.md — i.e. only a stylistic DRY hint, not content-wrong
  • MEDIUM if uncertain / context-dependent
  • DROP (strip from the report) if the hardcode is an explicit reference to the config (e.g. "respect restriction from athlete_static") — that is intended

Read the full file on GitHub · 192 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. 6d ago First seen · 192 lines · 82 tokens per session scan A 1fd9f3dbdc26

Subscribe to this mod's changes

config-auditor is an agent published in the GitHub repository airbone42/360-data-athlete (22 stars, last pushed today), licensed MIT. It adds 82 tokens to every session and 1,888 once invoked, about $0.0004 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 2026-08-30.