CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

dingtalk-chat

钉钉群聊与消息。Use when 收发/搜索消息、建群、群治理、Bot/Webhook、文件、AI/A2UI 卡片流式,或仅限 IM 的消息谓词筛选。跨源主题/行为轨迹走 dingtalk-aisearch;DING/班级群走 dingtalk-misc;邮件走 dingtalk-mail。

Quality

90%

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.

An exceptionally dense, highly actionable routing skill: concrete dws commands for every user endpoint, explicit validation and retry semantics, and a clean one-level reference bundle. Its weaknesses are mild — a duplicated policy block between the runtime contract and Shortcut sections, and a few inline sections that could be pushed to references.

Suggestions

De-duplicate the Schema/Help/unknown-command policy: state it once in 最小 DWS 执行契约 and have the Shortcut 发现 section reference it instead of repeating it verbatim.

Move the inline DWS_RUNTIME_CONTRACT policy detail into a shared reference (e.g. references/contracts.md, which already exists) and keep only a one-line pointer in the body.

Consider moving the dws-intent HTML comment markers out of the Golden Route table cells into a maintenance note or build artifact, since they consume context tokens without guiding runtime behavior.

DimensionReasoningScore

Conciseness

The body is telegraphic and rule-dense with essentially zero explanation of concepts Claude already knows — every line is a routing rule or boundary. However, the Schema/Help/unknown-command policy is stated twice in nearly verbatim form (in 最小 DWS 执行契约 and again in Shortcut 发现), and the dws-intent HTML comments add tokens, so minor trimming is possible. This is 'efficient with minor instances that could be trimmed' (anchor 4) rather than the every-token-earns-its-place ideal of anchor 5.

4 / 5

Actionability

The Golden Route table gives copy-paste-ready commands with flags and placeholders (e.g. `dws chat +dm --to <姓名> --content <内容>`, `dws chat +search-msg ... --start/--end`), and the schema inspection command is fully specified with --jq projection. Specific commands cover all common cases, matching the top anchor.

5 / 5

Workflow Clarity

Multi-step flows are explicitly sequenced with validation checkpoints: write lifecycle (校验目标/限制/内容/确认 → 串行执行 → 验证), group lifecycle shown as a chain (创建 → 保存 CID → dismiss), and a numbered 错误最短路径 with a bounded retry feedback loop (仅 retryable=true 重试一次) and confirmation gating. Destructive/batch operations all carry explicit validation (写后读回, 清理须确认、验证, 检查 complete/hasMore), so no cap applies.

5 / 5

Progressive Disclosure

The body is an overview/routing layer with a clearly signaled 按需加载 table mapping scenarios to one-level-deep references, and all 14 referenced bundle files (references/chat/*.md, references/card/*.md, references/*.md) exist as real files. Minor gaps keep it below anchor 5: the Help/Schema policy block is duplicated between the contract section and Shortcut 发现, and the inline DWS contract block is policy detail that could live in a shared reference.

4 / 5

Total

18

/

20

Passed

Description

92%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 dense, well-scoped description that explicitly answers both what and when, enumerates concrete capabilities, and disambiguates against sibling DingTalk skills. The only weakness is a few internal-jargon trigger terms (predicate filtering, A2UI) that users would not naturally say.

Suggestions

Replace or supplement jargon trigger terms like '消息谓词筛选' and 'AI/A2UI 卡片流式' with natural user phrasings such as '按条件筛选消息' and '卡片消息'.

Add common natural trigger terms users would actually say, e.g. 未读消息、@我的消息、发消息, to broaden keyword coverage.

DimensionReasoningScore

Specificity

"收发/搜索消息、建群、群治理、Bot/Webhook、文件、AI/A2UI 卡片流式,或仅限 IM 的消息谓词筛选" lists multiple concrete capabilities (send/receive/search messages, group creation, group governance, bots/webhooks, files, card streaming, predicate filtering) that comprehensively cover the DingTalk IM domain, matching the top anchor. Nothing reads as vague filler, and no neighboring anchor lists this many concrete actions.

5 / 5

Completeness

Both parts are explicit: the what is "钉钉群聊与消息" plus the capability list, and the when is an explicit "Use when 收发/搜索消息、建群..." clause with concrete trigger phrases. This matches the top anchor exactly; anchor 4 would require the 'when' to be less explicit than it is.

5 / 5

Trigger Term Quality

Natural terms users would say are present (消息, 群聊, 建群, 搜索消息, Bot, Webhook, 文件), giving good coverage, but terms like "消息谓词筛选" (message predicate filtering) and "A2UI 卡片流式" are internal jargon a user would not naturally say, and common variations like 未读/@我 are missing. This sits between anchor 3 and 5, noticeably above the midpoint but short of comprehensive natural-term coverage.

4 / 5

Distinctiveness Conflict Risk

The description actively routes away from sibling skills ("跨源主题/行为轨迹走 dingtalk-aisearch;DING/班级群走 dingtalk-misc;邮件走 dingtalk-mail"), establishing a clear IM-only niche with minimal conflict risk. This exceeds anchor 4's 'minor overlap risk' — boundaries are explicitly drawn.

5 / 5

Total

19

/

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: 10 deeper-than-1-level, 1 suspicious

Warning

referenced_paths_exist

Referenced path issues: 10 deeper-than-1-level

Warning

Total

12

/

16

Passed

Repository

DingTalk-Real-AI/dingtalk-workspace-cli

Path

skills/multi/dingtalk-chat/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.