CtrlK
BlogDocsLog inGet started
Tessl Logo

dingtalk-tag

钉钉数字员工的自然语言创建、查询、修改、发布、删除、能力资源、执行状态、本地 Profile 落盘与 DSH 接入。Use when 用户说创建或管理数字员工、改人设/岗位/响应模式、发布或询问下线能力、查执行状态或 trace、管理数字员工 Skill/MCP、把已有数字员工转换为本地 Profile,或接入本地 DSH。命令前缀:dws dingtalk-tag。

71

Quality

89%

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

86%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, high-signal instruction skill: dense domain constraints, explicit dry-run/confirmation checkpoints for risky operations, and exemplary one-level progressive disclosure to five real reference files. The main gaps are abbreviated (non-copy-paste) command forms and workflows expressed as constraints rather than sequenced steps.

Suggestions

Inline one or two complete, copy-paste-ready command examples (full flags for a common create and connect) so the skill body is executable without opening references.

Convert the unbind→connect and create-failure recovery rules into short numbered sequences with explicit success/failure checkpoints, keeping the constraint list for the rest.

DimensionReasoningScore

Conciseness

Every bullet is dense, domain-specific policy Claude cannot know (e.g., "digitalTagEmployeeProfile.type…仅支持 open_code、local_agent", "deviceId 是稳定设备标识,runtimeBindingId 是服务端绑定 ID") with zero padding and no explanation of known concepts. Not 4: there is no over-explanation anywhere to trim — each token earns its place.

5 / 5

Actionability

Gives concrete commands and flags ("dws dingtalk-tag connect --agent-uuid ... --channel dsh", "先用 --dry-run --format json 展示计划", "connect list/status/stop/restart") plus exact field names. Not 5: commands are abbreviated with '...' and alternatives like 'skill|mcp create|update|delete|list|query' rather than copy-paste-ready full invocations, and complete argument details are delegated to the reference files.

4 / 5

Workflow Clarity

Multi-step flows carry explicit checkpoints — "两步先 dry-run 并汇总确认", "旧机器 connect unbind 成功后,新机器…connect", "connect 失败后…检查 connect status 再恢复", "重试禁止再次执行 create" — including validation for destructive ops (delete 不可逆, confirmation requirements). Not 5: these are stated as constraint bullets rather than a sequenced workflow with feedback loops; the step-by-step sequences themselves live in the reference files, leaving some sequence assembly implicit.

4 / 5

Progressive Disclosure

The ~45-line body is a genuine overview: a routing table plus a purpose-signaled index ("先用 dingtalk-tag-index.md 路由;生命周期读取 manage.md…能力资源读取 capability.md;执行排障读取 run.md"), with all five referenced files present in references/ and only one level deep. Not 4: navigation is explicit (an index file with per-topic pointers), and no content that belongs in a reference file is inlined.

5 / 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 strong description: comprehensive concrete capability list plus an explicit, well-phrased 'Use when' trigger clause in the user's natural vocabulary. The only gap is modest synonym coverage in trigger terms.

Suggestions

Add a few synonym trigger terms (e.g., 数字人、数字员工下线、digital employee) to broaden natural-phrase matching.

DimensionReasoningScore

Specificity

Lists multiple concrete actions — "自然语言创建、查询、修改、发布、删除、能力资源、执行状态、本地 Profile 落盘与 DSH 接入" — giving comprehensive coverage of the skill's operations, matching the top anchor. Not 4 because coverage spans the full lifecycle (create/query/modify/publish/delete/capability/run-state/profile/connect) rather than having minor gaps.

5 / 5

Completeness

Explicitly answers both questions: the opening sentence states what the skill does (specific operations), and "Use when 用户说…" gives concrete trigger phrases. Not 4 because the 'when' clause is not merely present but enumerates concrete trigger scenarios, matching the 5 anchor exactly.

5 / 5

Trigger Term Quality

The "Use when 用户说创建或管理数字员工、改人设/岗位/响应模式、发布或询问下线能力、查执行状态或 trace、管理数字员工 Skill/MCP…" clause covers many natural user phrasings across intent variants. Not 5 because common synonyms/variants (e.g., 数字人、digital employee、English equivalents) are absent, though the coverage present is good.

4 / 5

Distinctiveness Conflict Risk

Clear niche (DingTalk 数字员工 lifecycle/DSH access) with distinct triggers and an explicit command prefix "dws dingtalk-tag" that disambiguates from sibling skills. Not 4: the overlap with generic 'dingtalk' skills is resolved by the narrow digital-employee scope and the stated prefix, leaving minimal conflict risk.

5 / 5

Total

19

/

20

Passed

Validation

81%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 13 / 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: 1 suspicious

Warning

Total

13

/

16

Passed

Repository
DingTalk-Real-AI/dingtalk-workspace-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.