CtrlK
BlogDocsLog inGet started
Tessl Logo

evolution-proposal

把 evolution-inbox 中的审计异常条目转化为可评审的进化提案(规则/技能/插件/路由补丁)。先做系统化问题分析(量化基线→全量聚合→根因分类→优先级排序),再用便宜模型起草、前沿模型评审、人工批准后走既有门禁固化。用于处理 evolution_scan.js 产出的异常条目、复盘高频反模式、把会话经验固化为规则或技能改进。不用于单轮小修改和 L0 治理规则的直接修改(只产提案不越权固化)。

66

Quality

80%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills/evolution-proposal/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

67%Weight 40%Scale 1-5

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

The body delivers an actionable, well-sequenced multi-stage workflow with concrete commands, named tools, validation checkpoints, and an acceptance checklist. Its main weakness is redundancy between the stage-4 template, the 输出契约 section, and the 验收 section, which inflates token cost without adding guidance.

Suggestions

Consolidate the stage-4 proposal output block and the '输出契约' section into a single source of truth for the required fields, replacing the duplicate field list with a pointer to the template.

De-duplicate the '验证' acceptance checklist against the stage-3 validation steps, or explicitly cross-reference them so the two do not restate the same criteria.

Flesh out the stage-3 review feedback loop as an explicit 'if REJECT/修改建议 → revise patch → re-run pre-check → re-submit' iteration so error recovery is unambiguous.

DimensionReasoningScore

Conciseness

The body is dense and operational with no basic-concept padding, but the '输出契约' section restates stage 4's proposal-template fields and the '验证' section overlaps earlier workflow content, so it could be meaningfully tightened; it is not merely minor trimming.

3 / 5

Actionability

Provides concrete commands (Get-Content .../inbox.jsonl, quality_report.py --strict, validate_repo.py --strict, dsh-config-sync --check, queue_cli_request/request_result, git diff --check), named analysis tools, and a copy-pasteable proposal template, with only minor argument gaps keeping it from fully executable.

4 / 5

Workflow Clarity

Five stages are clearly sequenced (stage 1 split into 1.1–1.7) with explicit validation in stage 3 (静态预检, 前沿模型评审, 成本自检), a 5-item 验收 checklist, and a status flow new→processing→applied|rejected; the REJECT→fix recovery loop is implied rather than fully iterated, a minor validation gap.

4 / 5

Progressive Disclosure

No bundle files exist, so the skill is a single well-sectioned document (核心规则/适用范围/工作流程 阶段1-5/输出契约/验证) with no nested references; the detailed proposal template and root-cause table could theoretically live in separate files but the organization is clean, leaving only minor gaps.

4 / 5

Total

15

/

20

Passed

Description

92%Weight 40%Scale 1-5

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 highly specific and self-bounding, naming a concrete pipeline (analyze→draft→review→approve→solidify) with explicit use-for and not-for guidance. Its only mild weakness is jargon density in the trigger phrasing, which keeps trigger-term coverage just below comprehensive.

DimensionReasoningScore

Specificity

Lists multiple concrete actions across the full pipeline — '转化为可评审的进化提案', '系统化问题分析(量化基线→全量聚合→根因分类→优先级排序)', '便宜模型起草、前沿模型评审、人工批准后走既有门禁固化', '复盘高频反模式' — giving comprehensive coverage rather than minor gaps.

5 / 5

Completeness

Clearly answers 'what' (convert→analyze→draft→review→approve→solidify) and explicitly answers 'when' via '用于处理...复盘...固化...' alongside a '不用于...' exclusion clause, matching the anchor for both what AND when with concrete trigger phrases.

5 / 5

Trigger Term Quality

Includes natural task-oriented phrases ('复盘高频反模式', '把会话经验固化为规则或技能改进') plus explicit 用于/不用于 clauses, but leans technical-jargon heavy (evolution-inbox, evolution_scan.js, L0 治理规则) without synonyms or extensions, so it falls just below comprehensive coverage.

4 / 5

Distinctiveness Conflict Risk

Occupies a clear niche (evolution-inbox audit anomalies → reviewable proposals) with an explicit boundary '只产提案不越权固化' and named depends_on skills, giving minimal conflict risk with other skills.

5 / 5

Total

19

/

20

Passed

Validation

93%

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

Validation15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

15

/

16

Passed

Repository
ooooooooooooooooooop/agent-tools
Reviewed

Table of Contents

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.