From 3-surgeons
Deep 3-phase multi-model analysis — cross-examine a claim using multiple AI models
npx claudepluginhub supportersimulator/3-surgeons --plugin 3-surgeonsThis skill uses the workspace's default tool permissions.
Cross-examination is the heavyweight tool. Use it when the stakes justify the cost and latency:
Generates design tokens/docs from CSS/Tailwind/styled-components codebases, audits visual consistency across 10 dimensions, detects AI slop in UI.
Records polished WebM UI demo videos of web apps using Playwright with cursor overlay, natural pacing, and three-phase scripting. Activates for demo, walkthrough, screen recording, or tutorial requests.
Delivers idiomatic Kotlin patterns for null safety, immutability, sealed classes, coroutines, Flows, extensions, DSL builders, and Gradle DSL. Use when writing, reviewing, refactoring, or designing Kotlin code.
Cross-examination is the heavyweight tool. Use it when the stakes justify the cost and latency:
Do NOT use for: trivial fixes, style changes, single-file edits with clear scope, questions with obvious answers.
3s cross-exam "Should we migrate from REST to GraphQL for the patient API?"
cross_examine(topic="Should we migrate from REST to GraphQL for the patient API?", depth="full")
The depth parameter defaults to "full" (3-phase). There is currently one depth level; future versions may support "quick".
Both the Cardiologist (GPT-4.1-mini) and Neurologist (Qwen3-4B) analyze the topic independently. Neither sees the other's work. This prevents anchoring bias -- each model forms its own opinion from scratch.
Each surgeon reviews the other's Phase 1 analysis. The Cardiologist reviews the Neurologist's work, looking for weaknesses, blind spots, and errors. The Neurologist does the same for the Cardiologist. This adversarial review is where most value is generated.
The Cardiologist (broader perspective from the external API model) synthesizes both analyses and both cross-reviews into a final report. The synthesis explicitly emphasizes disagreements -- where the surgeons diverge is the most valuable signal.
The result contains four key fields:
| Field | What It Tells You |
|---|---|
cardiologist_report | Initial analysis + cross-review of neurologist's work |
neurologist_report | Initial analysis + cross-review of cardiologist's work |
synthesis | Combined view emphasizing disagreements |
total_cost | USD spent on external API calls |
--- Cross-Review ---) highlights what the other missed. These blind spots are the highest-value findings.--- Cardiologist ---
[Initial analysis of the topic]
--- Cross-Review ---
[Cardiologist's critique of Neurologist's analysis]
--- Neurologist ---
[Initial analysis of the topic]
--- Cross-Review ---
[Neurologist's critique of Cardiologist's analysis]
--- Synthesis ---
[Combined findings, disagreements highlighted]
Cost: $0.0042 | Latency: 3200ms
If one surgeon is unavailable (endpoint down, timeout), the cross-examination still completes with the available surgeon. The result will have None for the missing surgeon's report. This is expected -- a one-surgeon analysis is still better than no analysis. Run 3s probe if a surgeon is persistently unavailable.
Every cross-examination is automatically logged to the evidence store (~/.3surgeons/evidence.db). This means:
If you need both surgeons' opinions but do NOT need cross-review or synthesis, use 3s consult "topic" instead. This runs Phase 1 only (independent analysis) at roughly 1/3 the cost and latency.