$npx -y skills add jmstar85/oh-my-githubcopilot --skill traceEvidence-driven tracing with competing hypotheses in parallel. Activate when user says: trace this, trace, why did this happen, root cause analysis, investigate why.
| 1 | # Trace |
| 2 | |
| 3 | Evidence-driven causal tracing using competing hypotheses. Use for ambiguous, causal, evidence-heavy questions where the goal is to explain WHY something happened. |
| 4 | |
| 5 | ## Good Entry Cases |
| 6 | - Runtime bugs and regressions |
| 7 | - Performance / latency behavior |
| 8 | - Architecture / premortem / postmortem analysis |
| 9 | - Config / routing / orchestration behavior |
| 10 | - "Given this output, trace back the likely causes" |
| 11 | |
| 12 | ## Core Contract |
| 13 | Always preserve: Observation → Hypotheses → Evidence For → Evidence Against → Best Explanation → Critical Unknown → Discriminating Probe |
| 14 | |
| 15 | ## Workflow |
| 16 | 1. Restate the observed result precisely |
| 17 | 2. Generate 3 deliberately different hypotheses: |
| 18 | - Code-path / implementation cause |
| 19 | - Config / environment / orchestration cause |
| 20 | - Measurement / artifact / assumption mismatch |
| 21 | 3. Assign @tracer to each hypothesis lane |
| 22 | 4. Each lane: evidence for, evidence against, critical unknown, discriminating probe |
| 23 | 5. Apply lenses: Systems, Premortem, Science |
| 24 | 6. Rebuttal round between top two hypotheses |
| 25 | 7. Rank, detect convergence, synthesize |
| 26 | |
| 27 | ## Output |
| 28 | ``` |
| 29 | ### Observed Result |
| 30 | [What happened] |
| 31 | |
| 32 | ### Ranked Hypotheses |
| 33 | | Rank | Hypothesis | Confidence | Evidence Strength | |
| 34 | |------|------------|------------|-------------------| |
| 35 | |
| 36 | ### Most Likely Explanation |
| 37 | [Current best explanation] |
| 38 | |
| 39 | ### Critical Unknown |
| 40 | [Single missing fact] |
| 41 | |
| 42 | ### Recommended Discriminating Probe |
| 43 | [Single next probe] |
| 44 | ``` |