self-evolution

Scope Guard + Process Evolution + Knowledge Evolution — 主动护栏与自我进化。 Use when: operator scope 发散偏离愿景、同类错误反复出现、SOP 流程缺口、有价值的知识/方法论值得沉淀。 Not for: 日常 SOP 推进（正常执行）、一次性个案 bug fix。 Output: Scope Guard Log 记录 / Evolution Proposal 提案 / Episode Card → Method/Skill 蒸馏 → Eval 验证。

Quality

81%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

77%Weight 40%Scale 1-3

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The content is highly actionable with concrete templates, file paths, and quantitative thresholds, and its workflows are clearly sequenced with validation checkpoints and feedback loops. The weaker dimensions are conciseness (cross-mode redundancy at ~254 lines) and progressive disclosure (referenced templates are not bundled and some detail could be split into separate files).

Suggestions

Dedupe the recurring guardrail/闭环 framing across the three modes into a single shared '硬护栏' section to reduce length and redundancy.

Bundle or inline-summarize the referenced templates (docs/episodes/TEMPLATE.md, docs/methods/TEMPLATE.md, evals/mode-c/TEMPLATE/) and ADR-015 so the progressive-disclosure references are verifiable and one level deep, or move the inline judge-dimension table and five-level ladder into referenced files.

Dimension	Reasoning	Score
Conciseness	The body is information-dense, uses tables efficiently, and assumes Claude's competence (references F167/F086/ADR-015/§16 without explanation), but at ~254 lines it repeats guardrail and闭环 concepts across the three modes and could be tightened.	2 / 3
Actionability	Provides concrete templates, specific file paths (docs/evolution-proposals/TEMPLATE.md, docs/episodes/TEMPLATE.md), and exact decision rules (≥2 evidence sources, 30-day replay, ≥3.5/5 overall, boundary ≥4/5, smoke/promotion gates), giving copy-ready executable guidance.	3 / 3
Workflow Clarity	Multi-step processes are explicitly sequenced with validation checkpoints — Mode B's 提案流程 (1–5 with commit/PR linkage and 30-day replay), Mode C's Episode→Distillation→Eval loop with A/B hygiene rules and pass thresholds — including feedback loops for error recovery.	3 / 3
Progressive Disclosure	Sections are well-organized and references to templates/ADRs are signaled inline, but the referenced template files and ADR-015 are not bundled (references/, scripts/, assets/ are absent), and substantial detail (judge dimension table, five-level ladder) is inline rather than split out.	2 / 3
	Total	10 / 12 Passed

Description

85%Weight 40%Scale 1-3

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is specific and complete, clearly stating what the skill does and when to use it with an explicit Not-for boundary. Its main weakness is trigger-term quality, which relies on internal jargon (operator scope, SOP, 愿景) rather than natural user-language keywords.

Suggestions

Add natural-language trigger variations a user would actually say (e.g. 'we keep making the same mistake', 'this drifted from the plan', 'worth remembering this method') alongside the current jargon.

Soften domain-internal terms (operator, SOP, 愿景) or pair each with a plain-language synonym so the description is router-friendly beyond the 三猫 team context.

Dimension	Reasoning	Score
Specificity	Names three concrete modes and specific output artifacts ('Scope Guard Log 记录', 'Evolution Proposal 提案', 'Episode Card → Method/Skill 蒸馏 → Eval 验证'), listing multiple concrete actions rather than vague language.	3 / 3
Completeness	Explicitly answers both 'what' (three modes + their outputs) and 'when' via a clear 'Use when:' clause plus a 'Not for:' clause, satisfying the explicit-trigger requirement.	3 / 3
Trigger Term Quality	Trigger phrases ('operator scope 发散偏离愿景', '同类错误反复出现', 'SOP 流程缺口') are relevant but lean toward internal jargon/abstract concepts and miss common natural-language variations a user would naturally say.	2 / 3
Distinctiveness Conflict Risk	Clear niche (self-evolution/scope-guard) with a 'Not for:' clause distinguishing it from routine SOP execution and one-off bug fixes, making wrong-skill triggering unlikely.	3 / 3
	Total	11 / 12 Passed

Validation

100%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: zts212653/clowder-ai
Path: cat-cafe-skills/self-evolution/SKILL.md
Commit: 80782c5

Reviewed: 11 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.