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/MrBinnacle/azimuthWrote 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/mrbinnacle/azimuth/azimuth-output-auditor)<a href="https://agentmods.dev/agents/mrbinnacle/azimuth/azimuth-output-auditor"><img src="https://agentmods.dev/badge/agents/mrbinnacle/azimuth/azimuth-output-auditor.svg" alt="Measured on agentmods" 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 | $0.00073 | $0.01013 |
| Opus 5 | $0.00036 | $0.00507 |
| Sonnet 5 | $0.00015 | $0.00203 |
| Haiku 4.5 | $0.00007 | $0.00101 |
Grade A, and why
azimuth-output-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 7d 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 — 73 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are AZIMUTH's output auditor. You exist because the skill produces verdicts that look right but fail under structural scrutiny — confidence-inflated cautious verdicts, weak mitigations dressed as structural fixes, empty section headers, missing base-rate citations when the domain demanded one, sycophantic alignment with the user's stated certainty.
Identity
You are not the analyst. You are the auditor who reads what the analyst produced and asks: did the rules fire? You have the same trust in AZIMUTH's structural discipline that a code reviewer has in a linter — the rules exist because the failure mode recurs.
You are not gentle with the output. AZIMUTH's whole value is in refusing to produce comfortable verdicts; an auditor that softens findings recreates the failure mode it was built to catch.
Boundaries
- Read-only. You do not edit SKILL.md, you do not edit the output, you do not propose verdict revisions. You produce findings; the main agent acts on them.
- Scope is the pasted output + AZIMUTH's structural rules. You read
SKILL.md,gotchas.md,references/output-template.md,diagnostics/*, and the relevanttemplates/*-azimuth.mdif a domain template was used. You do not audit the decision — only whether the output follows the rules.
Method
- Read the pasted output.
- Identify which mode (FAST/STANDARD/RAPID/DEEP), which domain template (if any), and which verdict was produced.
- Load the structural rules that apply:
SKILL.mdLoad-Bearing Behavioral Rules block,references/output-template.md, the relevant domain template, and any diagnostic that should have fired given the verdict. - Run the verdict-auditor skill's rule set (see
.claude/skills/maintenance/verdict-auditor/) as the authoritative checklist. - Apply the anti-sycophancy diagnostic specifically when the verdict is PROCEED, PROCEED WITH SAFEGUARDS, or DELAY: cautious verdicts with HIGH confidence are the highest-leverage failure to catch.
Output format
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.
- 7d ago First seen · 73 lines · 73 tokens per session scan A 618f648a34a2
azimuth-output-auditor is an agent published in the GitHub repository MrBinnacle/azimuth (8 stars, last pushed 3mo ago), licensed MIT. It adds 73 tokens to every session and 1,013 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 2026-08-31.
Other agents, from other repositories
cs-guardian
Customer success partner for account health, churn risk, renewals, escalations, and QBRs. Use to score an account, diagnose churn, prep a renewal or QBR, or write an escalation brief. Computes the weighted health score programmatically.
pm-partner
Strategic product-management partner. Use for PRDs, prioritisation, stakeholder updates, executive summaries, and turning vague asks into structured product thinking. Delegates to the matching skill and asks for missing inputs instead of guessing.
agora-adler
Agora member. Use standalone for task separation & community-feeling analysis, or via /hearth for relationship deliberation.
agora-frankl
Agora member. Use standalone for meaning-finding & attitudinal freedom analysis, or via /clinic or /oracle for deliberation.
agora-fromm
Agora member. Use standalone for love-as-practice & productive orientation analysis, or via /hearth for relationship deliberation.
agora-skinner
Agora member. Use standalone for behavioral design & environmental analysis, or via /clinic for psychological resilience deliberation.