CtrlK
BlogDocsLog inGet started
Tessl Logo

lark-apps

妙搭(Spark/Miaoda)应用开发与托管:应用创建、本地全栈开发、云端生成迭代、创意设计(UI mockup / 可交互原型 / 线框图 / 落地页 / 仪表盘 / 幻灯片 deck / 视觉探索)、AI相关能力和飞书平台能力或者其他外部能力集成、日志/Trace/监控指标/PV/UV 查询、环境变量管理、应用协作者与协作权限设置、应用角色与成员管理、自动化触发器(定时/记录变更/Webhook/飞书审批)。当用户要开发/新建一个系统·工具·平台·应用,或要本地开发 / 云端开发 / 修改 / 部署 / 发布 / 上线 / 拿可分享链接,或用 HTML 做页面·网站·部署到妙搭,或要设计 / design / mockup / prototype / wireframe / 做 PPT / deck / 视觉探索,或提到妙搭/Spark/Miaoda(应用运行时域名形如 *.aiforce.cloud)、应用数据库、应用文件存储、开放 API Key、可见范围、应用协作者/开发权限、应用角色/角色成员、线上日志、接口请求量、错误量、延迟、访问量、环境变量、给妙搭应用配自动化任务/定时触发/审批通过后自动触发时使用。不负责普通云盘文件上传(lark-drive)、飞书文档编辑(lark-doc)、原生幻灯片创建(lark-slides)。

74

Quality

93%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

88%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 strong operational skill body: an intent-routing table with real, resolvable reference files, executable commands throughout, and decision/guardrail sections with explicit validation gates and error-recovery loops for destructive operations. The main improvement areas are the fully inlined collaborator-permissions section and a repeatedly referenced creative-design/creative-design.md file that is missing from the bundle.

Suggestions

Move the detailed collaborator/permission rules (typed-ID requirements, feature_not_available / copy_download_by / external_invite edge cases) into a dedicated references/lark-apps-member.md and keep only the dry-run/--yes confirmation pattern inline.

Add the missing creative-design/creative-design.md to the bundle (it is referenced as the full development-and-publish flow for html/creative apps in five places), or repoint those links to an existing reference.

Replace the vague '对应 access-scope reference' cell with proper links to references/lark-apps-access-scope-get.md and references/lark-apps-access-scope-set.md, matching the style of every other routing row.

DimensionReasoningScore

Conciseness

The body is dense, routing-first prose that assumes Claude's competence and explains nothing Claude already knows, but the fully inlined ~25-line collaborator/permissions section (e.g. 'copy_download_by 也只在 +member-settings-get 的响应中读取…', 'feature_not_available(OpenAPI code 3340005)') and rules repeated across sections (app_id resolution, release handling) could be trimmed or moved out. This is anchor 4 ('efficient; minor instances… that could be trimmed'), not 5, because of that inline bulk.

4 / 5

Actionability

Commands are complete and copy-paste ready with exact flags and placeholders: 'lark-cli apps +member-add --app-id <app_id> --member-type openid --member-id ou_xxx --perm view --dry-run --as user', 'lark-cli apps +get --app-id <meta_token> -q '.data.app.app_id'', 'lark-cli auth login --domain apps'. Not 4: there are no missing key details in the common cases.

5 / 5

Workflow Clarity

Multi-step flows are explicitly sequenced with validation checkpoints: the '选择开发路径' decision table is ordered before intent routing, app_id resolution has ordered steps 1–3, destructive ops gate on '--dry-run' → user confirmation → '--yes', releases poll 'finished/failed' with PENDING-approval handling, and git failures loop through credential refresh → retry → stop and report. Destructive/batch operations all carry validation, so the ≤3 cap does not apply; feedback loops match the top anchor.

5 / 5

Progressive Disclosure

The intent-routing table maps 25+ intents to one-level-deep reference files and all 30 references/lark-apps-*.md links resolve, giving good structure. Kept at 4 rather than 5 because 'creative-design/creative-design.md' is referenced repeatedly but absent from the bundle, the access-scope row says '对应 access-scope reference' instead of a real link, and the member/collaborator rules are entirely inline with no reference file.

4 / 5

Total

18

/

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.

A model router-skill description: concrete capability inventory, an explicit use-when clause with abundant natural trigger terms in both Chinese and English, and explicit non-goals that prevent mis-triggering sibling skills. Its only trade-off is sheer length — the trigger enumeration is exhaustive to the point of density — but every phrase is load-bearing rather than padded.

DimensionReasoningScore

Specificity

The description enumerates many concrete capabilities — '应用创建、本地全栈开发、云端生成迭代、创意设计(UI mockup / 可交互原型 / 线框图 / 落地页 / 仪表盘 / 幻灯片 deck / 视觉探索)…日志/Trace/监控指标/PV/UV 查询、环境变量管理、应用协作者与协作权限设置、应用角色与成员管理、自动化触发器' — with comprehensive coverage and no vague filler, matching the top anchor rather than the 'minor gaps' anchor below it.

5 / 5

Completeness

It explicitly answers both 'what' (the enumerated capability list) and 'when' via the explicit trigger clause '当用户要开发/新建一个系统·工具·平台·应用…时使用', mirroring the model 5 example structure; a 4 would require a weaker or less explicit 'when' clause, which is not the case here.

5 / 5

Trigger Term Quality

It includes the natural phrases users would actually say — '部署 / 发布 / 上线 / 拿可分享链接', '做 PPT / deck', 'mockup / prototype / wireframe' — plus bilingual synonyms (妙搭/Spark/Miaoda) and the runtime domain pattern '*.aiforce.cloud', matching the comprehensive-synonyms anchor.

5 / 5

Distinctiveness Conflict Risk

It carves a clear niche (Miaoda/Spark app platform) and adds explicit exclusions — '不负责普通云盘文件上传(lark-drive)、飞书文档编辑(lark-doc)、原生幻灯片创建(lark-slides)' — which actively disambiguate from sibling skills, giving minimal conflict risk.

5 / 5

Total

20

/

20

Passed

Validation

75%

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

CriteriaDescriptionResult

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: 3 missing, 4 suspicious

Warning

Total

12

/

16

Passed

Repository
larksuite/cli
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.