CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

skill-creator-skill

当用户想要创建、编辑、改进或调试 OpenLoaf 自定义技能(Skill)时触发。典型说法:"帮我创建一个技能"、"做一个新 skill"、"把刚才的操作封装成技能"、"写一个能自动 XX 的技能"、"改一下这个 skill"、"这个 skill 为什么不触发"、"编辑我的自定义技能"、"加个全局技能"、"给当前项目加个技能"。任何涉及 `.openloaf/skills/` 目录下 `SKILL.md` 的创建 / 修改 / 调优请求都应加载本技能。也适用于用户想理解技能格式、排查触发问题、或把对话里的工作流固化成可复用能力的场景。不用于:内置技能(如 file-ops、email-ops 等)的修改——那些是平台随版本发布的只读能力。

72

Quality

91%

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

SKILL.md
Quality
Evals
Security

Quality

Content

81%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.

A well-structured instructional body: dense, platform-specific guidance with concrete templates, validation checkpoints, and an explicit iterate-and-retry loop. It misses the top tier mainly for lacking the end-to-end example it recommends for other skills, minor recap duplication, and no layered split of its longer template content.

Suggestions

Add one small end-to-end example SKILL.md (frontmatter + body) so the guide practices the '示例:1-2 个端到端完整示例' rule it prescribes — this would lift actionability to fully copy-paste-ready coverage.

Trim duplication between the 第X步 sections and the 铁律 recap (e.g., 'description 宁宽勿窄' and 'Glob 查冲突' each appear twice) — keep 铁律 as one-line pointers rather than restated guidance.

Move the recommended body-structure template and `openloaf.json` schema/colorIndex details into a short reference file (or explicitly justify keeping the single-file layout) to fully earn the progressive-disclosure top anchor.

DimensionReasoningScore

Conciseness

Largely efficient — tables, decision trees, and code blocks assume Claude's competence and teach platform-specific facts (paths, `openloaf.json`, colorIndex palette). Minor over-explanation remains: the 铁律 section restates several step-section points, and the colorIndex palette line is slightly padded. Not a 5 because some duplication between 第X步 sections and 铁律 could be trimmed; not a 3 because there is no teaching of concepts Claude already knows.

4 / 5

Actionability

Concrete and executable: exact write-path tables, a copy-paste `openloaf.json` example, a Glob conflict-check command, and a symptom/cause/fix troubleshooting table. Falls short of 5 because the skill's own recommended body structure includes a "示例:1-2 个端到端完整示例" section, yet the body itself contains no complete end-to-end example of a finished skill.

4 / 5

Workflow Clarity

Five numbered steps (理解需求 → 编写 → openloaf.json → 保存 → 验证与迭代) with explicit validation checkpoints (Glob 同名冲突检查 before writing; 开新对话测试触发 after) and a genuine feedback loop (排查表 → 修改 SKILL.md → 重试), plus a 铁律 checklist. Matches the top anchor.

5 / 5

Progressive Disclosure

Single-file skill with well-organized sections, clear headers, and no buried or nested references (no `references/`/`scripts/`/`assets/` bundle exists, so it is scored on the body itself). Good structure overall, but at ~205 lines of inline content it sits below the anchor-5 ideal of appropriately split content with well-signaled references (e.g., the recommended body template or `openloaf.json` schema could live in a reference file).

4 / 5

Total

17

/

20

Passed

Description

100%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.

An exemplary description: concrete action verbs, a rich set of quoted natural trigger phrases with colloquial variants, explicit what/when guidance, and a clear exclusion boundary. Anti-drift re-checks against the 4-level anchors confirm each dimension fits the 5 anchor best.

DimensionReasoningScore

Specificity

Lists multiple concrete actions — "创建、编辑、改进或调试" plus "理解技能格式、排查触发问题、把对话里的工作流固化成可复用能力" — covering the domain comprehensively with no vague filler.

5 / 5

Completeness

Explicitly answers both what ("创建、编辑、改进或调试 OpenLoaf 自定义技能") and when ("当用户想要…时触发" plus enumerated typical sayings and scenarios), with a concrete exclusion boundary ("不用于:内置技能…").

5 / 5

Trigger Term Quality

Nine quoted, natural user phrasings including colloquial variants ("帮我创建一个技能", "改一下这个 skill", "加个全局技能", "把刚才的操作封装成技能") plus technical markers (`.openloaf/skills/`, `SKILL.md`) — comprehensive synonym and phrasing coverage.

5 / 5

Distinctiveness Conflict Risk

Clear niche (OpenLoaf custom skills under `.openloaf/skills/` with distinct quoted triggers) and an explicit "不用于" clause excluding built-in skills, minimizing overlap with other skills.

5 / 5

Total

20

/

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.

Validation — 15 / 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

itsablabla/openloaf-web

Path

apps/server/src/ai/builtin-skills/skill-creator/SKILL.md

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.