From 3-surgeons
Confidence-weighted vote from multiple surgeons to validate claims and check assumptions
How this skill is triggered — by the user, by Claude, or both
Slash command
/3-surgeons:consensusThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Consensus is the lightweight decision tool -- a quick confidence-weighted vote rather than a full cross-examination. Use it when:
Consensus is the lightweight decision tool -- a quick confidence-weighted vote rather than a full cross-examination. Use it when:
Do NOT use for: complex architectural decisions (use cross-examination), health checks (use probe), or risk assessment (use sentinel).
3s consensus "Redis LPUSH is O(1) for single elements"
consensus(claim="Redis LPUSH is O(1) for single elements")
Each surgeon receives the claim and responds with a JSON assessment:
{
"confidence": 0.85,
"assessment": "agree",
"reasoning": "Redis LPUSH is documented as O(1) per element..."
}
The system then calculates a weighted consensus score:
weighted_score = sum(confidence_i * assessment_value_i) / sum(confidence_i)
Where assessment values are: agree = +1.0, uncertain = 0.0, disagree = -1.0.
| Score Range | Meaning | Action |
|---|---|---|
| +0.7 to +1.0 | Strong agreement | Proceed with confidence. Both surgeons agree the claim is valid. |
| +0.3 to +0.7 | Moderate agreement | Proceed but verify. One surgeon is less confident or uncertain. |
| -0.3 to +0.3 | Uncertain / Split | Do not proceed blindly. Consider escalating to full cross-examination. |
| -0.7 to -0.3 | Moderate disagreement | The claim is likely wrong or oversimplified. Investigate further. |
| -1.0 to -0.7 | Strong disagreement | The claim is rejected by both surgeons. Re-examine your assumption. |
The result includes each surgeon's individual vote:
Cardiologist: agree (confidence=0.90)
Neurologist: uncertain (confidence=0.60)
Weighted score: +0.50
Total cost: $0.0008
Key things to notice:
Escalate from consensus to full cross-exam when:
Do NOT escalate when:
Consensus makes 2 LLM calls (one per surgeon). The Neurologist call is free (local model). The Cardiologist call costs approximately $0.0004-0.001 depending on claim length. This is roughly 1/5 the cost of a full cross-examination.
# Step 1: Quick consensus check
3s consensus "Switching from SQLite to PostgreSQL is safe for our <100 user base"
# Result: weighted_score=+0.45 (moderate, but uncertain zone edge)
# Step 2: Score is borderline -- escalate
3s cross-exam "Should we switch from SQLite to PostgreSQL given <100 concurrent users?"
# Result: Full analysis reveals Neurologist flagged WAL mode contention
# that Cardiologist missed, while Cardiologist noted migration complexity
The consensus caught that this was not a clear-cut decision. The cross-examination revealed exactly why.
npx claudepluginhub supportersimulator/3-surgeons --plugin 3-surgeonsProvides UI/UX resources: 50+ styles, color palettes, font pairings, guidelines, charts for web/mobile across React, Next.js, Vue, Svelte, Tailwind, React Native, Flutter. Aids planning, building, reviewing interfaces.
Fetches up-to-date documentation from Context7 for libraries and frameworks like React, Next.js, Prisma. Use for setup questions, API references, and code examples.