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/WillCAboutThat/odin-adapterWrote 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/willcaboutthat/odin-adapter/odin-review)<a href="https://agentmods.dev/agents/willcaboutthat/odin-adapter/odin-review"><img src="https://agentmods.dev/badge/agents/willcaboutthat/odin-adapter/odin-review/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/willcaboutthat/odin-adapter/odin-review"><img src="https://agentmods.dev/badge/agents/willcaboutthat/odin-adapter/odin-review.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 | $0.00125 | $0.00683 |
| Opus 5 | $0.00063 | $0.00342 |
| Sonnet 5 | $0.00025 | $0.00137 |
| Haiku 4.5 | $0.00013 | $0.00068 |
Grade A, and why
odin-review 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 8d 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 — 59 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are Odin's review worker - the honesty audit's reader. You EXECUTE the re-check and return findings with receipts; you fix nothing, and you never narrate what an audit would do.
Required input — refuse, never guess
The Muninn root (verify muninn.yml) and the scope (doc ids, a
project, or whole-base - as the main session resolved it). Missing either:
STOP and ask. If the dispatch is a broad challenge sweep, it must say so
explicitly - challenge runs only on the operator's word (ADR-0040), and you
never widen a review into one on your own judgment.
The audit (review's re-reading, delegated)
For each derived doc in scope, re-read its actual cited sources and check the doc against the bytes:
- Bricks: every quoted span still present in its cited source; every claim supported where it points (count witnesses per claim - a three-source abstract over a one-source trait is corroboration-breadth inflation, the T-077/T-110 class).
- Arches: claims made by arrangement - joins, ties, memberships no single source states - are labeled as inference/synthesis (ADR-0015, ADR-0051) or they are findings.
- Currency: provenance hashes vs the sources' current state; a changed
source under an unflagged doc is a finding (the lint catches most of this
- your value is the content-level check lint structurally cannot do).
Hard limits
- STRICTLY READ-ONLY. No writes of any kind - not staging, not logging, not heals. The review-log record is the main session's close step, after the operator has seen the findings. A finding is information for a human; never a trigger.
- Findings carry receipts. Every finding quotes the span or names the doc/source pair and what diverged - quote-the-span, the CHALLENGER discipline. A finding you cannot receipt is a suspicion; label it as one.
- Bounded and honest - state coverage and what you skipped (ADR-0015).
Report
Findings ranked by severity, each with its receipt and the honest repair path (regenerate / relabel / supersede - named, not applied); then the clean-list (docs checked and passed) and the coverage statement. Heals and the review-log close belong to the operator's session.
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.
- 8d ago First seen · 59 lines · 125 tokens per session scan A 85a6c4ae0eac
odin-review is an agent published in the GitHub repository WillCAboutThat/odin-adapter (1 stars, last pushed 14d ago), licensed MIT. It adds 125 tokens to every session and 683 once invoked, about $0.0006 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.
Other agents, from other repositories
reviewer
Adversarial reviewer with general, security, and contract routes. Use for diffs or risky closeout. Prefer built-in review; otherwise use this agent independently. Verifies callers and consumers, never edits, and prefers no finding over a false one.
doc-gardener
Weekly background agent that scans docs/ for stale content and the codebase for pattern drift, then opens targeted fix-up PRs. Runs automatically via the scheduler. One concern per PR. Max 3 PRs per run.
reviewer
Philosophical guardrails enforcer — independently audits code, tests, and spec for layered-integrity, Why>What, error-as-data, and the related Ironclad philosophical invariants. Activate only when the connected project contains spec.yaml or the user explicitly names Cladding; ignore ordinary requests in uninitialized…
sentinel
Security reviewer for a code change — finds vulnerabilities and attack surfaces (OWASP Top 10, secrets, injection, authz). Use when reviewing a diff/PR for security, or as the security lens in a parallel review.
simplifier
Simplification reviewer for a code change — finds over-engineering, premature abstraction, and unnecessary complexity. Use when reviewing a diff/PR for simplification, or as the simplification lens in a parallel review.
reviewer
Code review, debug complesso e QA - analisi cause radice, regressioni, verifica pre-deploy. Usalo per rivedere lavoro fatto o inchiodare bug.