メニュー
Confidence-based scoring system for review findings — calibrates issue severity with evidence strength to filter false positives. Use when reviewing code, analyzing findings, or scoring issues.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Demo pipeline state machine — 7-phase autonomous sequence with delegation context templates, phase transition logic, BLOCKED recovery strategies, and demo-state.json schema. Used exclusively by demo-conductor.
Cinematic narration style guide for demo-conductor — ANSI-coloured banner formats, live pipeline scoreboard, audience-facing language, phase summaries, and error narration patterns. Keeps the demo presentation-quality throughout.
Standardized completion and escalation protocol for subagent responses. Ensures the conductor can machine-parse every subagent return. Use when reporting completion status back to the orchestrator.
Cross-session learnings lifecycle — schema, storage, retrieval, and pruning of lessons learned during orchestrator sessions. Use when managing learnings via the /learn command.
Agent Teams assembly and task injection — selects appropriate team, validates prerequisites, estimates cost, injects tasks into the shared task list, and manages team lifecycle.
Token and cost tracking with model tier enforcement
| name | confidence-scoring |
| description | Confidence-based scoring system for review findings — calibrates issue severity with evidence strength to filter false positives. Use when reviewing code, analyzing findings, or scoring issues. |
| user-invocable | false |
Rate every finding 0-100 based on evidence strength:
| Score | Level | Meaning | Action |
|---|---|---|---|
| 0-24 | Not confident | Likely false positive, insufficient evidence | Filter out |
| 25-49 | Somewhat confident | Might be real, needs investigation | Note only |
| 50-74 | Moderately confident | Real but minor, or real but context-dependent | Include if reviewer |
| 75-89 | Highly confident | Real and important, strong evidence | Always include |
| 90-100 | Certain | Verified issue with definitive evidence | Flag as critical |
80 — only findings scored ≥80 are reported to the user by default. Configurable in sdlc-config.md.
For each finding, evaluate:
### Finding: {description}
- **Confidence:** {score}/100
- **Evidence:** {exact code reference}
- **Category:** Bug | Security | Convention | Performance | Maintainability
- **Recommendation:** {specific fix}