CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

dingtalk-shared

钉钉(DingTalk) MultiSkill 的轻量共享入口。Use when 用户泛称 DWS/钉钉操作但未明确产品、请求跨产品编排、需要 URL 类型预检或产品边界消歧。清晰的单产品操作优先使用对应 dingtalk-* 子 skill;本 skill 只提供全局执行契约和按需 reference 导航,不承载产品命令全集。

Quality

86%

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 lean, highly operational shared-entry SKILL.md: dense runtime contract, explicit routing order, and error handling with validation checkpoints and retry semantics. The only gaps are mechanics deferred to references without a named entry command, and a few orphan bundle files not discoverable from the entry table.

Suggestions

Name the actual node-type probe command (or a one-line example) in routing step 1 for the alidocs /i/nodes/ flow, so the URL resolution chain is executable without first opening url-patterns.md.

Add 渐进加载 table rows (or a one-line pointer) for channel-login.md, runtime-contract.md and recipes/meta.md so every bundle file is discoverable from the entry point.

DimensionReasoningScore

Conciseness

The ~80-line body is dense and operational throughout: the runtime contract bullets, the situation→file table, and the routing/error sequences contain zero padding and no explanations of concepts Claude already knows. Every token carries a rule, a command, or a routing decision, matching the 'lean and efficient; every token earns its place' anchor.

5 / 5

Actionability

Guidance is concrete and executable: exact flags and commands (`--format json`, `dws auth login --intl --device`, `dws doc info`), concrete JSON fields (`result.fileId`, `linkSourceInfo.nodeId`, `isOrgCurrent=true`, `retryable`, `retry_after_seconds`), and a specific multi-hop URL resolution chain. It is not 5 because key mechanics are delegated to references (e.g. "先执行节点类型探测" names no command, and leaf-Schema lookup procedure lives in schema-usage.md), leaving minor gaps for a cold start.

4 / 5

Workflow Clarity

Multi-step processes are explicitly sequenced with validation checkpoints: a 5-step routing order ending in "向用户追问,不要猜测", a 5-step error shortest-path with retry caps ("最多重试一次"), retry gated on "明确 retryable=true", a confirmation gate ("确认后才加 --yes"), and mandatory post-write verification ("写后验证结果,不凭退出码宣称成功"). This matches the top anchor including feedback loops and stop conditions.

5 / 5

Progressive Disclosure

The body is a clear entry overview with a well-signaled situation→must-read-file table, and all eleven referenced bundle files exist and are one level deep. It is not 5 because three bundle files (references/channel-login.md, references/runtime-contract.md, references/recipes/meta.md) are not reachable from any entry-point link, and the cross-skill link ../dingtalk-misc/references/profile.md is unverifiable within this bundle — minor organization gaps.

4 / 5

Total

18

/

20

Passed

Description

83%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 description that clearly and explicitly answers both what the skill does and when to use it, including useful negative triggers that route single-product work elsewhere. The main weaknesses are meta-level phrasing of capabilities and some system jargon (e.g. 产品边界消歧) that users would not naturally say.

Suggestions

Phrase one or two triggers in natural user vocabulary (e.g. "用户说『用钉钉/用 DWS 查一下』但没说查文档还是表格") alongside the existing jargon terms, to strengthen trigger_term_quality.

Add one concrete example of what the entry resolves (e.g. "将泛称请求消歧为 doc/drive/wiki/aitable/calendar 等具体产品") to raise specificity from router-level to capability-level description.

DimensionReasoningScore

Specificity

The description names the domain (钉钉/DingTalk DWS) and lists several concrete functions — "轻量共享入口", "全局执行契约和按需 reference 导航", "URL 类型预检", "产品边界消歧" — matching the 'lists several specific actions with minor gaps' anchor. It falls short of 5 because the functions are described at a router/meta level without enumerating more granular capabilities.

4 / 5

Completeness

Both questions are explicitly answered: what ("本 skill 只提供全局执行契约和按需 reference 导航,不承载产品命令全集") and when ("Use when 用户泛称 DWS/钉钉操作但未明确产品、请求跨产品编排、需要 URL 类型预检或产品边界消歧"), with a concrete negative trigger ("清晰的单产品操作优先使用对应 dingtalk-* 子 skill"). This matches the top anchor exactly.

5 / 5

Trigger Term Quality

Natural keywords users would say are present: "泛称 DWS/钉钉操作", "DingTalk", "跨产品编排". Not 5 because phrases like "产品边界消歧" and "URL 类型预检" are system jargon users would rarely say, and product-level synonyms that lead users here indirectly are not covered.

4 / 5

Distinctiveness Conflict Risk

The skill is an umbrella over the dingtalk-* family but actively manages overlap: it explicitly redirects clear single-product requests to the corresponding dingtalk-* sub-skill and reserves itself for vague or cross-product requests. Minor overlap risk with the sibling product skills remains by design, so it is not a 5.

4 / 5

Total

17

/

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

relative_links

Relative link issues: 2 deeper-than-1-level, 1 suspicious

Warning

referenced_paths_exist

Referenced path issues: 2 deeper-than-1-level

Warning

Total

12

/

16

Passed

Repository

DingTalk-Real-AI/dingtalk-workspace-cli

Path

skills/multi/dingtalk-shared/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.