CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

investigation-first

调查研究:在下判断、做决策、给建议之前,先用第一手材料弄清现状,产出「事实 / 推断 / 未知」三栏调查结论。当上下文不完整、证据薄弱、领域陌生、用户说"先别动手"、或你发现自己在凭印象作答时触发;已充分调查后的执行阶段不要重复触发。 English: Investigate before judging. Gather firsthand facts and produce a fact / inference / unknown ledger before any recommendation. Trigger when context is incomplete, evidence is thin, the domain is unfamiliar, the user says "look first", or you notice yourself answering from impression; skip when investigation is already done and the task is pure execution.

Quality

85%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

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 tight, highly actionable investigation methodology with a clear sequenced workflow and built-in validation gates. The only notable gap is a broken external reference (original-texts.md is cited but absent).

DimensionReasoningScore

Conciseness

The body is lean and assumes competence — the 用/不用 list, 6-step 操作规程, and fill-in template all earn their place — but the epigraph quote and the opening '实事求是' framing line are minor trimmable padding that keeps it just below anchor 5.

4 / 5

Actionability

Highly actionable for an instruction-only skill: concrete steps naming specific sources (git log, blame, source/config/data, run, logs) plus a copy-paste-ready output template; per the scoring notes, absence of code is not penalized when guidance is this concrete.

5 / 5

Workflow Clarity

A clear numbered 6-step sequence with explicit validation gates — the 'unknown' column acts as a stop-gate ('禁止在未知栏非空且关系到结论时下定论'), and the '改判断不改事实' rule plus budget discipline provide feedback loops and a checklist matching anchor 5.

5 / 5

Progressive Disclosure

Well-organized into clear sections (用/不用, 操作规程, 输出模板, 纪律, 交接) with the supplementary source material pushed to a separate file, but the referenced original-texts.md does not exist in the bundle, so the one external reference is a dead link — a minor gap below anchor 5.

4 / 5

Total

18

/

20

Passed

Description

76%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, well-structured description that clearly states both capability and trigger conditions including a skip clause. Its main weakness is trigger-term quality: most triggers are internal-state descriptions rather than phrases users naturally say.

Suggestions

Add natural user-uttered trigger phrases users would actually say, e.g. 'investigate first', 'check before deciding', 'look before you leap', alongside the existing 'look first'.

Surface a concrete trigger keyword or two earlier in the description so the natural-language signal leads rather than the internal-state descriptions.

Consider naming the output artifact ('fact / inference / unknown ledger') even more prominently as the distinguishing capability to sharpen distinctiveness from generic research skills.

DimensionReasoningScore

Specificity

Names several concrete actions — 'Gather firsthand facts', 'produce a fact / inference / unknown ledger', 'Investigate before judging' — with a distinctive three-column output artifact, but stops short of the multi-verb comprehensive list of anchor 5.

4 / 5

Completeness

Explicitly answers both what (investigate, gather firsthand facts, produce a three-column ledger before any recommendation) and when (concrete trigger phrases plus a 'skip when' clause), matching the anchor for clearly answering both with concrete triggers.

5 / 5

Trigger Term Quality

Includes one clean user-uttered phrase ('the user says "look first"') but most triggers describe Claude's internal state ('context is incomplete', 'evidence is thin', 'answering from impression') rather than natural user utterances, and common synonyms like 'check first' or 'look before you leap' are missing.

3 / 5

Distinctiveness Conflict Risk

The fact/inference/unknown ledger and 'investigate before judging' framing carve a fairly distinct niche with an explicit skip clause, but 'investigation' broadly still overlaps with general research/exploration skills, keeping it just below anchor 5.

4 / 5

Total

16

/

20

Passed

Validation

93%

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

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

relative_links

Relative link issues: 1 missing

Warning

Total

15

/

16

Passed

Repository

HughYau/qiushi-skill

Path

skills/investigation-first/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.