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.
npx skills add valpere/chorus --skill councilgit clone --depth 1 https://github.com/valpere/chorusWrote 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/skills/valpere/chorus/council)<a href="https://agentmods.dev/skills/valpere/chorus/council"><img src="https://agentmods.dev/badge/skills/valpere/chorus/council/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/skills/valpere/chorus/council"><img src="https://agentmods.dev/badge/skills/valpere/chorus/council.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 | $0.00062 | $0.00670 |
| Opus 5.5 | $0.00025 | $0.00268 |
| Sonnet 5.5 | $0.00012 | $0.00134 |
| Haiku 4.5 | $0.00006 | $0.00067 |
Grade A, and why
chorus-council 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 — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Chorus: LLM Council
When to use
- User asks for a multi-agent perspective on a decision or problem
- User says "council", "multiple opinions", "ask all agents", "LLM council"
Your role in the council
You are the SCOPE reviewer. Focus on: unnecessary complexity, premature abstractions, whether the smallest viable solution was chosen.
Agent roles
| Agent | Focus |
|---|---|
| Claude | Correctness — logic errors, type safety, security issues |
| Gemini | Edge cases — unusual inputs, failure modes, alternatives |
| Cursor | Integration — codebase fit, patterns, dependency risks |
| Kilo | Maintainability — readability, naming, long-term tech debt |
| (you) | Scope — unnecessary complexity, smallest viable solution |
Invocation
Spawn all four agents in parallel, then add your own scope review:
claude --print "You are the CORRECTNESS reviewer in an LLM council. Focus on: logic errors, type safety, off-by-one bugs, security issues. Be concise — bullet points.\n\nTask: <task>" --dangerously-skip-permissions &
CLAUDE_PID=$!
gemini --prompt "You are the EDGE-CASES reviewer in an LLM council. Focus on: unusual inputs, failure modes, alternatives not considered. Be concise — bullet points.\n\nTask: <task>" --yolo --output-format text &
GEMINI_PID=$!
agent -p --force "You are the INTEGRATION reviewer in an LLM council. Focus on: codebase fit, consistency with existing patterns, dependency implications. Be concise — bullet points.\n\nTask: <task>" &
CURSOR_PID=$!
kilo run --auto "You are the MAINTAINABILITY reviewer in an LLM council. Focus on: readability, naming clarity, long-term tech debt. Be concise — bullet points.\n\nTask: <task>" &
KILO_PID=$!
wait $CLAUDE_PID $GEMINI_PID $CURSOR_PID $KILO_PID
Graceful degradation: Check each agent with command -v <binary> before spawning. Skip missing agents and warn the user. Proceed as long as at least 2 agents (including yourself) are available.
Output handling
After collecting outputs (fewer if some agents were skipped) plus your own scope review, synthesize as chairman:
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 · 62 lines · 62 tokens per session scan A 6b1d83fdccdb
chorus-council is a skill published in the GitHub repository valpere/chorus (10 stars, last pushed 5mo ago), licensed MIT. It adds 62 tokens to every session and 670 once invoked, about $0.0002 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 2026-08-31.
Other skills, from other repositories
dead-code-cleanup
Finds and removes dead or orphaned code, tests, and Storybook stories, including trivial tests that prove nothing. Validates that code is truly unused before deleting anything. Trigger on: "remove dead code", "find unused exports", "what code can we delete", "prune the repo".
comment-reaper
Finds and removes comments that add nothing: ones that restate the code, cite issue numbers, narrate refactor history, or pad a one-liner with JSDoc. Keeps the WHY comments. Trigger on: "reap comments", "remove unnecessary comments", "too many comments", "comment audit".
pr-review-fix-pipeline
Reviews a PR and applies fixes for the non-controversial findings: review, fix, test, and commit in one pass. Writes code, so use it only on feature branches where automatic edits are fine. Trigger on: "review and fix PR", "auto-fix the PR", "review PR #X and fix issues", "run the review pipeline".
code-health-audit
Runs the "embarrassing code" / overengineering sweep and turns tech-debt anxiety into a short list of validated simplifications. Syncs the base branch and checks the worktree first so it never audits stale code. Trigger on: "code health audit", "overengineering audit", "what can we simplify", "tech debt sweep".
abstraction-finder
Finds recurring patterns worth extracting into shared code and proposes the smallest form for each (helper, util, base class) with the call sites it would replace. Enforces the rule of three. Trigger on: "where are we repeating ourselves", "find duplication", "what could be abstracted", "could this be DRY'd up".
band-aid-finder
Finds fixes that hide a symptom instead of fixing the cause, traces each to its root cause, and proposes the real fix. Allows deliberate, documented mitigations. Trigger on: "is this a real fix or a band-aid", "am I treating the symptom", "why does this keep happening", "find the root cause".