Tessl • The management layer for AI agent skills

An exemplary lean skill body: executable commands, an explicit research workflow with a stated call-priority rule, and a concrete template, all with zero filler. The single notable gap is the absence of a validation/verification checkpoint for the generated skill.

Suggestions

Add a short verification step at the end of the workflow (e.g., smoke-test one generated command with lark-cli schema or a dry-run, and confirm the new description's trigger clause is unambiguous) to close the workflow_clarity gap.

Optionally note what to do when the 调研 API step fails (beyond delegating to lark-openapi-explorer), giving the workflow an explicit error-recovery branch.

Dimension	Reasoning	Score
Conciseness	The body is lean and command-first: CLI capability summary, a numbered research procedure, a fill-in template, and four one-line principles. No concept explanations Claude already knows, no padding — every token earns its place. Fits the top anchor; not 4 because there is no over-explanation anywhere to trim.	5 / 5
Actionability	Provides copy-paste-ready executable commands ("lark-cli api GET /open-apis/vc/v1/rooms --params '{\"page_size\":\"50\"}'", "lark-cli schema <service.resource.method>", "lark-cli auth login --domain <name>") and a concrete SKILL.md template whose placeholders ('xxx', '{...}') are appropriate for a skill-generator. Common cases are covered; fits the top anchor.	5 / 5
Workflow Clarity	A clear research sequence is present (--help → schema → api 裸调 → 确定 API、参数和 scope), followed by the template — sequence and priorities are explicit ("优先级：Shortcut > 已注册 API > api 裸调"). Not 5 because there is no validation checkpoint or feedback loop for the produced skill (e.g., verify the generated commands run or the description triggers correctly); not 3 because the sequence itself is coherent, ordered, and the skill is non-destructive so the batch/destructive cap does not apply.	4 / 5
Progressive Disclosure	The body (~75 lines) is appropriately inlined and well-sectioned, with one-level, clearly signaled references to sibling skills ("[lark-openapi-explorer](../lark-openapi-explorer/SKILL.md)" and the template's pointer to ../lark-shared/SKILL.md). No bundle files exist or are needed; navigation is easy. Fits the top anchor for a compact, well-organized skill with one-level references.	5 / 5
	Total	19 / 20 Passed

Description

82%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.

A strong, compact description: it states a concrete capability, includes an explicit and natural trigger clause, and occupies a well-delineated niche. Its only real weakness is modest action coverage — two named modes of use rather than a fuller enumeration.

Suggestions

Enumerate one or two additional concrete capabilities (e.g., 查询参数定义/schema、添加权限 scope) to lift specificity from 3 to 4-5.

Add a few common trigger synonyms or alternate phrasings (e.g., 'Lark', 'Feishu', 'make a lark-cli skill') to broaden trigger-term coverage.

Dimension	Reasoning	Score
Specificity	Names the domain ("创建 lark-cli 的自定义 Skill") and two concrete action types ("包装原子 API 或编排多步流程"), but coverage of actions is not comprehensive — no other capabilities are enumerated. Fits anchor 3 (domain plus 1-2 concrete actions) rather than 4, which requires several specific actions with only minor gaps.	3 / 5
Completeness	Explicitly answers both: what ("创建 lark-cli 的自定义 Skill") and when ("当用户需要把飞书 API 操作封装成可复用的 Skill…时使用") with concrete trigger phrases. The trigger clause is the direct equivalent of "Use when…", matching the top anchor; not 4 because the 'when' is already fully explicit and specific.	5 / 5
Trigger Term Quality	Contains natural, likely-user phrasing — "飞书 API", "lark-cli", "封装", "可复用", "编排多步流程" — that a user requesting this skill would plausibly say. Not 5 because some common variations and synonyms (e.g., English "Lark", "Feishu", "make/create a skill") are missing.	4 / 5
Distinctiveness Conflict Risk	Occupies a clear niche (authoring lark-cli skills that wrap Feishu APIs) with distinct triggers tied to lark-cli/飞书 terminology; minimal overlap risk with other skills. Fits the top anchor cleanly.	5 / 5
	Total	17 / 20 Passed

Validation

75%

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 — 12 / 16 Passed

Validation for skill structure

Criteria	Description	Result
metadata_version	'metadata.version' is missing	Warning
metadata_field	'metadata' should map string keys to string values	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
relative_links	Relative link issues: 2 suspicious	Warning

	Total	12 / 16 Passed

Repository: larksuite/cli
Path: skills/lark-skill-maker/SKILL.md
Revision: 05ac58f

Reviewed: 1 month ago

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.

lark-skill-maker

Table of Contents

Is this your skill?