CtrlK
BlogDocsLog inGet started
Tessl Logo

research

用真实检索工具查找可追溯的事实、数据和来源。

57

Quality

65%

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

Fix and improve this skill with Tessl

tessl review fix ./chapter10/multi-role-transfer/skills/research/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

80%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.

The body is admirably concise and gives concrete, actionable instructions with a clear handoff model, but the workflow lacks explicit validation checkpoints before recording findings or handing off. Adding one or two verify-then-proceed steps would raise workflow clarity without sacrificing brevity.

Suggestions

Insert an explicit validation checkpoint, e.g. '记录前确认数值/单位/URL 来自检索结果而非记忆' (before recording, confirm the value/unit/URL came from results, not memory), turning the existing guardrail into a verify step.

Add a one-line worked example of a `web_search` query and the verbatim record format to lift actionability from 4 to 5.

Make the handoff condition explicit and ordered (e.g. 'if computation is needed → data_analysis; else if code execution → coding; else → writing') so the decision branch is unambiguous.

DimensionReasoningScore

Conciseness

The ~6-line body is lean and efficient: it assumes Claude's competence, assigns a role and concrete tool without explaining what web_search or a URL is, and every line earns its place, matching 'lean and efficient; every token earns its place'.

5 / 5

Actionability

Gives concrete, specific guidance — name the tool (`web_search`), prefer primary/authoritative sources, record numbers/units/time-ranges/URLs verbatim, and explicit handoff targets (`data_analysis`/`coding`/`writing`) — but provides no worked example queries, fitting 'mostly executable guidance; minor gaps' rather than the fully copy-paste level 5.

4 / 5

Workflow Clarity

A sequence is present (search → record verbatim → decide handoff) but there are no explicit validation checkpoints (e.g. confirm a source is authoritative before recording, or verify a number was actually found), matching 'steps listed but validation gaps; checkpoints missing or implicit'.

3 / 5

Progressive Disclosure

This is a simple, single-purpose skill under 50 lines with no bundle files and no need for external references; its self-contained, well-organized content qualifies for the simple-skills exception that lets progressive disclosure score 5.

5 / 5

Total

17

/

20

Passed

Description

50%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.

The description clearly states what the skill does but omits any explicit trigger ('Use when...') guidance and uses fairly generic keywords, leaving it anchored at the mid-level across all dimensions. Adding concrete invocation triggers and a few natural synonyms would lift specificity, completeness, and distinctiveness together.

Suggestions

Add an explicit 'when' clause, e.g. '当用户要求查找可追溯的事实、数据或来源时使用' (Use when the user asks for traceable facts, data, or sources).

Include natural user-facing synonyms and surface terms (e.g. '搜索' search, '查资料' look up, '引用' citation) alongside '检索' to improve trigger term quality.

Sharpen distinctiveness by naming the concrete tool and output (e.g. '用 web_search 检索并带回带 URL 的可引用来源') so it is less likely to collide with general research skills.

DimensionReasoningScore

Specificity

Names the domain (retrieval) and one concrete action ('查找可追溯的事实、数据和来源' — find traceable facts, data, sources) but does not list multiple specific actions, matching the '1-2 concrete actions, not comprehensive' anchor rather than the multi-action level 4.

3 / 5

Completeness

The 'what' is clear (use real retrieval tools to find traceable facts/data/sources), but there is no 'Use when...' clause or equivalent trigger guidance for when to invoke the skill, which per the guidelines caps completeness at 3.

3 / 5

Trigger Term Quality

Contains relevant keywords ('事实' facts, '数据' data, '来源' sources, '检索' search) a user might say, but offers no synonyms or specific surface terms and leans on the somewhat technical '检索工具', fitting 'some relevant keywords but missing common variations'.

3 / 5

Distinctiveness Conflict Risk

The traceable-facts/sources niche is somewhat specific and distinguishable, but the generic trigger language ('事实、数据、来源') could overlap with other research-adjacent skills, matching 'somewhat specific but could still overlap'.

3 / 5

Total

12

/

20

Passed

Validation

100%

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

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
bojieli/ai-agent-book
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.