llm-council
Skill1 starsChanged 6 months agoMITView on GitHub
Run a question through 5 AI advisors who analyze independently, peer-review anonymously, then synthesize a verdict. Invoke for strategic decisions with genuine uncertainty where multiple perspectives add value.
54 other projects use a version of this.
Adopted by
Most recently taken up by
- seifsg/llm-counciltheir own version, first seen Nov 15, 2026
- contactandrewchl-wq/turtle-mcptheir own version, first seen Nov 15, 2026
- harshitsinghbhandari/domain-expansiontheir own version, first seen Nov 8, 2026
- Pawan028/llm-counciltheir own version, first seen Nov 8, 2026
- yusuf787x/claude-skillstheir own version, first seen Nov 1, 2026
- ataidecarlos/agentictheir own version, first seen Nov 1, 2026
Used with
Projects with it also use graphify, premortem and 7 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
IgorGanapolskySkillgraphify
Query the local Graphify-Labs AST knowledge graph for ThumbGate architecture questions (what connects X to Y, explain a symbol, PreToolUse/gates/adapters). Slash: /graphify.
Also in 6 projects that use this
neurofooSkillpremortem
Pre-mortem analysis that imagines a plan has failed, then works backward to identify causes and preventions. Use before launches, major decisions, or risky initiatives to surface hidden risks.
Also in 3 projects that use this
AnasX7Skillbrandkit
Premium brand-kit image generation skill for creating high-end brand-guidelines boards, logo systems, identity decks, and visual-world presentations. Trained for minimalist, cinematic, editorial, dark-tech, luxury, cultural, security, gaming, developer-tool, and consumer-app brand systems. Optimized for intentional logo concepting, refined composition, sparse typography, strong symbolic meaning, premium mockups, art-directed imagery, and flexible grid layouts.
Also in 3 projects that use this
AnasX7Skillminimalist-ui
Clean editorial-style interfaces. Warm monochrome palette, typographic contrast, flat bento grids, muted pastels. No gradients, no heavy shadows.
Also in 3 projects that use this
Microsoft AzureSkillhumanizer
Tone enforcement patterns for external-facing community responses
Also in 3 projects that use this
SentrySkillskill-creator
Create new agent skills following the Agent Skills specification. Use when asked to "create a skill", "add a new skill", "write a skill", "make a skill", "build a skill", or scaffold a new skill with SKILL.md. Guides through requirements, writing, registration, and verification.
Also in 6 projects that use this
- +1 more from AnasX7
Other versions54 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 54
- rohitg00/pro-workflow★ 2.9ksame name
- aiwithremy/claude-skills-llm-council★ 1.8ksame name
- am-will/codex-skills★ 1ksame name
- gcpdev/llm-council-skill★ 428same name
- ffroliva/gflow-cli★ 396same name
- happycapy-ai/Happycapy-skills★ 137same name
- nicmarti/skills-weaver★ 30same name
- oliwoodman/llm-council-skill★ 25same name
- gestordeaudiencia/llm-council-skill-ptbr★ 20same name
- am-will/llm-council★ 16same name
44 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository13 other files it gives its agents.
- agent-cliSkill
- askSkill
- blueprintSkill
- brainstormSkill
- carveSkill
- dispatchSkill
- handoffSkill
- improve-claude-mdSkill
- leanSkill
- orientSkill
- react-auditSkill
- shipSkill
- tddSkill
The file
We split this skill into the rules it gives an agent
5 rules. 0 are common, 5 are distinctive, and 0 are very specific to this project.
0
Common
Most of the 48 edited versions we read keep these.
5
Distinctive
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 16
Not for: factual lookups, creation tasks, or questions with one right answer.
- Distinctive: line 37
Add key context and constraints from the user
- Distinctive: line 40
If the question is too vague to council, ask one clarifying question.
- Distinctive: line 50
All 5 must run in parallel.
- Distinctive: line 68
Reasoning approach: Before writing the verdict, reason silently through:
“Common” is measured: kept by at least 6 in 10 of the 48 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
LLM Council
Multi-perspective analysis based on Andrej Karpathy's LLM Council methodology. Five independent advisors, anonymous peer review, one synthesized verdict.
When to Invoke
- Strategic decisions where being wrong is expensive
- Trade-offs where reasonable people genuinely disagree
- High-stakes choices where blind spots could be costly
Not for: factual lookups, creation tasks, or questions with one right answer.
The Advisors
| Advisor | Orientation |
| Contrarian | Finds fatal flaws and hidden risks |
| First Principles | Strips to fundamentals, questions assumptions |
| Expansionist | Finds upside everyone else is missing |
| Outsider | Zero domain context — fresh eyes |
| Executor | What do you actually do Monday morning? |
Full persona descriptions and prompt templates: Council Playbook
Process
Step 1 — Frame the Question
Take the user's raw question and reframe it as a clear, neutral prompt:
- State the core decision or question
- Add key context and constraints from the user
- Identify what's at stake — why this decision matters
If the question is too vague to council, ask one clarifying question. Then proceed.
Step 2 — Independent Analysis (5 parallel sub-agents)
Spawn all 5 advisors simultaneously using the Agent tool. Recommended model for advisors: sonnet-4.5 (cost/quality balance — each persona is narrow and short-form; Opus×5 in parallel is ~5× cost for marginal gain). Each advisor receives:
- Their persona from the Council Playbook
- The framed question
- Instruction: respond in 150-300 words, don't hedge, lean fully into your assigned angle
All 5 must run in parallel. Sequential spawning lets earlier responses bleed into later ones.
Step 3 — Peer Review (5 parallel sub-agents)
Collect all 5 responses. Anonymize them as Response A through E — randomize the mapping so there's no positional bias.
Spawn 5 new sub-agents in parallel. Recommended model for reviewers: sonnet-4.5 (structural comparison across 5 short texts — well within Sonnet's range). Each reviewer sees all 5 anonymized responses and answers:
- Which response is the strongest and why? (pick one)
- Which response has the biggest blind spot?
- What did ALL responses miss?
Each review: under 200 words.
Step 4 — Chairman Synthesis + Output
Spawn one Chairman sub-agent. Recommended model: opus-4.7 with extended thinking (high effort) — 11-input synthesis is the cognitive bottleneck of this skill. The Chairman receives everything de-anonymized: original question, all 5 advisor responses, all 5 peer reviews.
Reasoning approach: Before writing the verdict, reason silently through:
- Which points appear in ≥3 advisor responses (convergence signal)?
- Where do reviewers disagree about the strongest response (genuine clash)?
- Which blind spots emerged ONLY in peer review, not in original advisors?
Identify at least 2 non-obvious findings before structuring output. Do not stream a first draft — produce a single considered verdict.
Produces the verdict in this structure:
- Where the Council Agrees — convergence points (high-confidence signals)
- Where the Council Clashes — genuine disagreements with both sides
- Blind Spots the Council Caught — emerged only through peer review
- The Recommendation — a clear answer, not "it depends"
- The One Thing to Do First — one concrete next step
The chairman can disagree with the majority if the dissenter's reasoning is strongest.
Display the full verdict directly to the user.
Output
- Verdict: displayed in conversation immediately after synthesis
- Transcript: saved to
./plans/council-{topic}-{YYYY-MM-DD}.mdwith: original question, framed question, all 5 advisor responses, all 5 peer reviews (with anonymization mapping revealed), and the chairman's full synthesis
Acceptance Checklist
- Question framed and confirmed with user
- All 5 advisors produced independent analyses (parallel)
- Peer review completed with anonymization (parallel)
- Chairman synthesis delivered with clear verdict
- Full transcript saved to
./plans/
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

