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/Avyayalaya/agent-councilWrote 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/commands/avyayalaya/agent-council/council-review)<a href="https://agentmods.dev/commands/avyayalaya/agent-council/council-review"><img src="https://agentmods.dev/badge/commands/avyayalaya/agent-council/council-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/commands/avyayalaya/agent-council/council-review"><img src="https://agentmods.dev/badge/commands/avyayalaya/agent-council/council-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 | $0.00029 | $0.00438 |
| Opus 5 | $0.00015 | $0.00219 |
| Sonnet 5 | $0.00006 | $0.00088 |
| Haiku 4.5 | $0.00003 | $0.00044 |
Grade A, and why
council-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 12d 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.
What it actually says
Council Review
Run the Agent Council against a text artifact. Five role-conditioned deliberators (Skeptic, Voice & Identity, Evidence & Calibration, Strategy & Stakes, Adjudicator) review the artifact in a 2-round async protocol with cross-read rebuttal, then synthesize a single verdict.
Pre-conditions
Before running this command, confirm:
- The
agent-councilPython package is installed (pip install agent-councilorpip install -e <local-clone>). - A
council.yamlexists in the working directory or a parent. If not, copy fromcouncil.yaml.exampleand edit:- Set
runtime.typeto your installed LLM CLI (claude_cli,lmstudio,ollama,mock_cli). - Point
context_refsat your voice corpus and goals doc (templates inexamples/).
- Set
Invocation
If the user mentions a file path: run python -m agent_council review <path> --tier=1 --config=council.yaml.
If they say "this file" or "current artifact" without naming one: ask which file they mean. Do not assume.
Always run with --tier=1 unless the user explicitly says otherwise — that's the tier this command is for.
Output
Print the verdict (SHIP / REVISE / HOLD), the adjudicator's reasoning, and each deliberator's specific concerns. If REVISE, surface the revision brief as a numbered list the user can act on.
If the verdict is HOLD, do NOT propose fixes — the Council blocked for a reason; surface the blockers and let the user decide.
If the verdict is SHIP with concerns, list the concerns inline but make clear the artifact passed.
A full structured verdict has been appended to council_log.jsonl. Mention this once for the user's records.
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.
- 12d ago First seen · 35 lines · 29 tokens per session scan A 9d92637c4a47
council-review is a command published in the GitHub repository Avyayalaya/agent-council (10 stars, last pushed 1mo ago), licensed MIT. It adds 29 tokens to every session and 438 once invoked, about $0.0001 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 commands, from other repositories
review
A guided command for reviewing code changes by examining the current Git diff, the record of edits between versions. It checks quality, bugs, security, performance, and maintainability.
CLAUDE
Multi-agent collaboration platform for persistent, proactive AI agents across rooms, workspaces, skills, and external services.
cc-council
Comprehensive multi-agent council review with 6 protocols, 10 specialists, scoped scoring (per-scope thresholds and weights), state machine orchestration, auto-fix, and 50+ configuration flags.
speckit.critique.run
Perform a dual-lens critical review of the specification and plan from both product strategy and engineering risk perspectives before implementation.
deep-review
Multi-pass deep code review — security, quality, and test gaps.
agent
Create and manage custom AI agents.