CtrlK
BlogDocsLog inGet started
Tessl Logo

dingtalk-aisearch

AI搜问:人员语义搜索与跨源定位。Use when 按姓名/工号/部门/职责/上下级或手机号线索找人,跨文档/消息/邮件/听记检索,或回溯“我发过/收到过”。完整手机号反查走 dingtalk-contact;找到 userId 后由 contact 补详情。命令前缀:dws aisearch。

72

Quality

92%

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

85%Weight 40%Scale 1-3

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

The body is highly actionable with well-sequenced SOPs and a clean progressive-disclosure structure pointing to real reference files. The only weakness is redundancy between the intent table, SOP-1, and the '高频硬约束' section, which inflates the token budget without adding new guidance.

Suggestions

Collapse '高频硬约束' into SOP-1 (or keep only the constraints not already stated there) to remove the near-duplicate rules on 完整保真, retry-once, contact handoff, multi-candidate, and --format json.

Trim the '意图表' to user-phrasing examples only and let the SOPs own the command mapping, avoiding re-stating the same triggers twice.

Merge the standalone '跨产品协作' bullets into SOP-1 step 4 (衔接), which already lists the same contact/chat/misc routing, to eliminate the second copy.

DimensionReasoningScore

Conciseness

The body is mostly lean and assumes Claude's competence, but '高频硬约束' largely restates SOP-1's rules (完整保真, 换维度重试一次, 切 contact 补详情, 多候选询问, --format json) and the '意图表' repeats the SOP triggers, so it could be tightened. Not level 1 (no concept explaining / padding) and not level 3 (redundancy means not every token earns its place).

2 / 3

Actionability

Commands are concrete and copy-paste ready with explicit flags and dimensions — e.g. 'dws aisearch person --keyword "<完整值>" --dimension <维度> --format json' and 'dws contact user get --ids <userId> --format json' — matching the fully-executable anchor.

3 / 3

Workflow Clarity

SOPs are numbered with '必须' markers, an explicit '失败' step (retry one dimension once, then 如实告知), multi-candidate handling, and a parse-then-act validation loop via --format json — matching the clear-sequence-with-checkpoints anchor.

3 / 3

Progressive Disclosure

SKILL.md is an overview with well-signaled one-level-deep references — '[aisearch.md](references/aisearch.md)', intent-guide.md, lite-recipes.md, ding.md — all of which exist as real files; content is appropriately split, matching the top anchor.

3 / 3

Total

11

/

12

Passed

Description

100%Weight 40%Scale 1-3

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is specific, trigger-rich, and self-routing: it states concrete capabilities, gives natural 'Use when' triggers, and disambiguates against sibling skills via explicit handoff rules. No weak dimensions; nothing to cap.

DimensionReasoningScore

Specificity

Lists multiple concrete actions — '人员语义搜索与跨源定位', '跨文档/消息/邮件/听记检索', and '回溯我发过/收到过' — matching the multiple-specific-actions anchor; not the level below which only names a domain plus a few actions.

3 / 3

Completeness

Explicitly answers both what ('人员语义搜索与跨源定位') and when ('Use when 按姓名/工号...找人...或回溯...') with an explicit trigger clause, matching the top anchor.

3 / 3

Trigger Term Quality

Natural user phrasing is well covered — '按姓名/工号/部门/职责/上下级或手机号线索找人', '谁负责', '我发过/收到过' — terms a user would actually say; richer than the 'some relevant keywords' level-2 anchor.

3 / 3

Distinctiveness Conflict Risk

Clear niche with distinct routing — '完整手机号反查走 dingtalk-contact', '找到 userId 后由 contact 补详情', '命令前缀:dws aisearch' — unlikely to trigger the wrong skill, matching the clear-niche anchor.

3 / 3

Total

12

/

12

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.

Validation12 / 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

referenced_paths_exist

Referenced path issues: 2 missing

Warning

Total

12

/

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.