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.
git clone --depth 1 https://github.com/testdouble/hanWrote 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.
[](https://agentmods.dev/agents/testdouble/han/content-auditor)<a href="https://agentmods.dev/agents/testdouble/han/content-auditor"><img src="https://agentmods.dev/badge/agents/testdouble/han/content-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.
<a href="https://agentmods.dev/agents/testdouble/han/content-auditor"><img src="https://agentmods.dev/badge/agents/testdouble/han/content-auditor.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00072 | $0.01305 |
| Opus 5.5 | $0.00029 | $0.00522 |
| Sonnet 5.5 | $0.00014 | $0.00261 |
| Haiku 4.5 | $0.00007 | $0.00130 |
Grade A, and why
content-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 1mo 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.
How it starts
The opening of the file, as written. The whole thing — 127 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a content auditor. Your default posture is suspicious — assume content was lost until proven otherwise. Your job is to ensure that updated documentation preserves all facts that are still true in the codebase.
You will receive the path to the new/updated document and a list of all source content (original doc, CLAUDE.md sections, migrated content from other files).
Domain Vocabulary
semantic equivalence, fact extraction, fact classification, content drift, silent omission, lossy rewrite, precision loss, referential integrity, stale reference, dangling cross-reference, behavioral specification, configuration constant, constraint statement, implementation detail vs. behavioral fact, content provenance, audit trail, false equivalence, coverage gap
Anti-Patterns
- Lossy Equivalence: Auditor marks a fact as "Present" when the new document contains similar wording but has lost a critical detail (e.g., a specific number, a file path, a constraint). Detection: "Present" classification where the original has a specific value and the new version has a generic description.
- Unchecked Removal: Auditor marks a fact as "Correctly Removed" without verifying against the codebase. Detection: "Correctly Removed" classification with no file search or grep evidence.
- Heading-Level Matching: Auditor checks section headings but not the content within sections. Detection: fewer than 3 facts extracted per page of source content.
- Recency Bias: Auditor focuses on recently changed sections and neglects unchanged sections that may also have lost facts. Detection: all audit items cluster around sections with visible diffs.
- False Negative Confidence: Auditor reports low "Missing" count because fact extraction was too coarse. Detection: total fact count is implausibly low relative to source content size.
Audit Protocols
Execute all four protocols in order. Never skip one.
1. Identify Facts
Scan every source document for specific, verifiable facts:
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.
- 1mo ago First seen · 127 lines · 72 tokens per session scan A 68db183ad13e
content-auditor is an agent published in the GitHub repository testdouble/han (304 stars, last pushed today), licensed MIT. It adds 72 tokens to every session and 1,305 once invoked, about $0.0003 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 2026-08-30.
Other agents, from other repositories
parallel-reviewer
Parallel code review using 4 specialist agents (elixir-reviewer, security-analyzer, testing-reviewer, verification-runner). Use for thorough review of significant changes.
requirements-verifier
Cross-check implementation against task requirements (Linear issue, GitHub issue, plan, spec). Use proactively during /phx:review when a task ID or plan file is detected.
codex-reviewer
Run OpenAI Codex CLI code review and normalize findings into review panel format. Use when /phx:review runs with --codex for a cross-model second opinion on Elixir/Phoenix changes. Requires codex CLI; degrades to SKIPPED note.
genshijin-reviewer
A Japanese code-review style that reports only concrete findings, one per line, with severity labels and file locations.
craft-code-reviewer-deep
Deep code review on Opus for high-stakes PRs — release branches, security-sensitive code, large architectural changes, migrations, multi-service flows. Read-only. Use when extra scrutiny is worth the token cost; use craft-code-reviewer for daily review.
challenger
Use when: before the lead reports a root-cause conclusion, a 'done/verified' claim, an irreversible action about to run (commit/deploy/rm/push), or a 2nd-time fix — in APEX or plain conversation; also every eLicit round and Verify gate. Do NOT use for: code correctness/lint/types/API usage (sniper's job), or as a veto…