Azimuth Output Auditor — Research agent for Claude Code
Stress-tests a pasted AZIMUTH analysis output against the skill's own structural rules.
How to install Azimuth Output Auditor
Installs to ~/.claude/agents/mrbinnacle-azimuth-azimuth-output-auditor.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/MrBinnacle/azimuth/HEAD/.claude/agents/azimuth-output-auditor.md -o ~/.claude/agents/mrbinnacle-azimuth-azimuth-output-auditor.md Restart Claude Code, or start a new session, for it to be picked up.
What Azimuth Output Auditor does
name: azimuth-output-auditor description: Stress-tests a pasted AZIMUTH analysis output against the skill's own structural rules. Use when a real AZIMUTH verdict has been produced and needs adversarial review for sycophancy, confidence inflation on cautious verdicts, output discipline failures, or hook-rule violations. Read-only — never modifies the skill or the output. tools: Read, Glob, Grep, Bash model: sonnet
You are AZIMUTH's output auditor. You exist because the skill produces ver
Alternatives in Research
- CLI Explore Agent — Read-only code exploration via Bash + CLI semantic dual-source analysis, with schema-validated structured outp 530 ★
- Strategist Critic — Empirical strategy critic and gatekeeper 250 ★
- Jogyo — Scientific research agent with Python REPL and structured output markers 244 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Loki
Devil's advocate that challenges assumptions, stress-tests designs, finds edge cases, and performs adversarial
Citation Auditor
Audits citations in a legal memo draft against the research files that produced it. Verifies every normative/c
Research Critic
Adversarial reviewer for external-research outputs produced by docs-researcher. Re-verifies claims against pri
Deep Research Critic
Adversarial reader of a nav-deep-research draft. Finds ignored counter-evidence, unsupported claims, verbatim
Consensus Synthesizer
Reads all validator reports, applies consensus-synthesis voting rules, handles disagreements via consensus-dis
Research Manager
Debate judge for Upbit 8-stage pipeline — weighs bull vs bear rounds and emits the 5-tier structured verdict J
Related Skills
Ralph Output Adversarial Review
Adversarially reviews the artifacts produced by a RALPH run (code/tests/UI/config) against repo policy (testin
Sop Stress Test
Claude skill that stress-tests SOPs, runbooks and checklists across business ops, cybersecurity and AI operati
Adversarial Test
Post-build adversarial review — a fresh agent stress-tests recent changes for bugs, regressions, conflicts, an