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）。

Quality

81%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

77%

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

A strong, highly actionable routing skill with clear sequenced workflows and validation gates for risky/destructive operations. The weak point is progressive disclosure: a repeatedly-cited creative-design.md reference is missing from the bundle, and one routing row leaves its reference as vague prose instead of a link to an existing file.

Suggestions

Add the missing creative-design/creative-design.md bundle file (or fix the path) — it is referenced as the full development+publish flow for html/creative apps, so its absence leaves a critical path dangling.

Turn the access-scope routing row into an explicit link to the existing references/lark-apps-access-scope-get.md and references/lark-apps-access-scope-set.md instead of the vague '对应 access-scope reference'.

Tighten the dense routing-table prose and de-duplicate guidance that appears in both the table and the 高频路径/护栏 sections to improve token efficiency.

Dimension	Reasoning	Score
Conciseness	Largely efficient and assumes competence (no explaining what git/env vars are), but the dense Chinese routing table and several long guideline sentences are padded and could be tightened; some rows carry near-duplicate detail already stated elsewhere.	2 / 3
Actionability	Provides concrete executable commands throughout (lark-cli apps +list --keyword, +get --app-id <app_id> -q '.data.app.app_id', git credential refresh, env-set --yes), with specific flags and exact query paths rather than vague direction.	3 / 3
Workflow Clarity	Multi-step flows are explicitly sequenced with validation checkpoints: app_id resolution has a numbered priority list, the publish path requires commit+push then +release-create/+release-get polling, and destructive ops (+env-delete, role delete) have confirmation gates and a 'stop and report' feedback loop on git-credential failures.	3 / 3
Progressive Disclosure	The intent-routing table cleanly points to one-level-deep reference files (mostly real and present under references/), but two navigation paths are broken/vague: creative-design/creative-design.md is referenced repeatedly yet the bundle file/directory does not exist, and the access-scope row says '对应 access-scope reference' with no link even though the two files exist.	2 / 3
	Total	10 / 12 Passed

Description

85%

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 highly specific, well-scoped description that clearly states capabilities, explicit trigger conditions, and out-of-scope boundaries. The main weakness is verbosity: the very long synonym-stuffed trigger lists hurt trigger-term clarity and token efficiency without adding much discovery signal.

Suggestions

Trim the dense synonym lists in the '当用户要…时使用' triggers to the handful of phrases users actually say, to reduce padding and improve trigger clarity.

Consider front-loading the single most representative 'Use when...' clause rather than burying it in a comma-heavy enumeration.

Keep the explicit out-of-scope exclusions (lark-drive/lark-doc/lark-slides) — they are the strongest distinctiveness signal and should stay prominent.

Dimension	Reasoning	Score
Specificity	Lists many concrete actions (应用创建、本地全栈开发、云端生成迭代、创意设计、日志/Trace/监控指标查询、环境变量管理、应用角色与成员管理、自动化触发器), each tied to specific sub-capabilities rather than vague language.	3 / 3
Completeness	Both 'what' (full capability catalog) and 'when' are explicit via the repeated '当用户要…或…时使用' trigger clauses, plus explicit out-of-scope exclusions (lark-drive/lark-doc/lark-slides).	3 / 3
Trigger Term Quality	It enumerates abundant natural terms (开发/新建、部署/发布/上线、设计/design/mockup/prototype/wireframe/PPT/deck), but the sheer density and the long lists of synonyms dilute the truly common phrasings a user would naturally say; not the cleanest coverage.	2 / 3
Distinctiveness Conflict Risk	Niche is sharply bounded to 妙搭/Spark/Miaoda (*.aiforce.cloud) with explicit exclusions of sibling skills, making conflict with other skills unlikely despite overlap risk in its own family.	3 / 3
	Total	11 / 12 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: 3 missing, 4 suspicious	Warning

	Total	12 / 16 Passed

Repository: larksuite/cli
Path: skills/lark-apps/SKILL.md
Commit: a7865cd

Reviewed: 1 day 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.