majiayu000/spellbook / skills/capability-distill
capability-distill
Verified能力蒸馏工作流——从用户批准的强模型访谈或真实任务轨迹中提取非显然的判断规则,形成可审计的 judgment packet,再交给 skill-audit 和 skill-creator 决定是否落成可加载 skill。当用户说“蒸馏这个模型的判断力”“把这次任务的关键决策固化下来”“模型窗口要关了,保留它在某类场景的判断”时使用。普通流程文档、直接写 SKILL.md、泛化最佳实践或未授权的会话日志扫描不使用本 skill。
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 10.1 KB |
| evals/evals.json | config | 4.1 KB |
| evals/files/incident_existing_skills.md | markdown | 759 B |
| evals/files/incident_timeline.md | markdown | 1.4 KB |
| evals/files/review_judgment_packet_v1.md | markdown | 1.1 KB |
| evals/files/review_trajectory.md | markdown | 1.3 KB |
evals/files/review_trajectory.md
View on GitHub# Redacted code-review trajectory
The repository and contributor identities are intentionally omitted.
1. A PR added a merge gate with a free-form `review_source` string and treated any non-empty value as valid.
2. The reviewer compared three options: document expected values, accept unknown values with a warning, or validate a closed set and fail closed.
3. Evidence showed that a misspelled source already passed the fixture. Because the field controlled merge authorization, the reviewer rejected the documentation-only and warning options.
4. The implementation changed the field to a closed set, treated missing and unknown values as blocked, and added a schema-valid negative fixture for the misspelled value.
5. The reviewer then noticed that self-review could still pass without a recorded reviewer-lane failure. The same fail-closed rule was extended to require that prerequisite evidence.
6. Focused tests and the full gate suite passed; the negative fixtures failed for the intended policy reason rather than for malformed schema.
7. Existing `security-review` guidance already owned generic fail-closed advice, while the repository's merge-gate skill owned the concrete review-source and prerequisite rules.
Outcome: the existing merge-gate skill was updated; no new generic security skill was created.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-analysis-optimizationmorluto/reaOptimize REA providers and shared analysis workflows for agent context, latency, CPU, and memory while preserving correctness and Evidence. Use for analysis bottlenecks and cache or query design; CI/test-suite optimization belongs to rea-ci-tests-audit. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| rea-ci-tests-auditmorluto/reaAudit REA CI and test suites for correctness, fixture drift, false passes, and measured performance; use for suite or workflow investigations and efficiency changes, rather than ordinary feature testing or a single test failure. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-auditDietrichGebert/ponytailQuality audit of the whole repo: bugs, security, real load, missing tests, speed, and what to delete. Most important first. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |