CtrlK
BlogDocsLog inGet started
Tessl Logo

investigation-first

触发:当你准备下判断、做决策或提出建议,但事实、上下文或一手信息还不充分时优先调用;常见信号包括 unknowns、信息缺口、证据不足、领域陌生、需要先摸清现状。 English: Trigger before making claims or decisions when context is incomplete, evidence is weak, or the domain is unfamiliar. Use this skill to investigate first, gather firsthand facts, and let reality shape the conclusion.

58

Quality

66%

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

Fix and improve this skill with Tessl

tessl review fix ./skills/investigation-first/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

50%

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

The content is a well-organized methodology skill with a clear sequencing of steps and a concrete operating protocol, but it is padded with repeated quotations, lacks validation checkpoints, and references a bundle file (original-texts.md) that is not present.

Suggestions

Trim the redundant Mao quotations and repeated 'investigate-before-concluding' restatements so each token earns its place and conciseness improves.

Add an explicit validation/verification checkpoint to the 操作规程 (e.g., 'confirm every checklist item is answered and every assumption is backed by a first-hand fact before writing the conclusion').

Either supply the referenced original-texts.md or remove the dangling '详细原著依据见 original-texts.md' pointer so progressive disclosure is consistent.

DimensionReasoningScore

Conciseness

The body is mostly efficient conceptual guidance rather than padding Claude's knowledge, but it is noticeably extended by repeated Mao quotations and restatements of the same 'investigate before concluding' principle, matching the 'Mostly efficient but includes some unnecessary explanation or could be tightened' anchor.

2 / 3

Actionability

It mixes a concrete 操作规程 (named steps, checkbox checklist, fixed output template) with abstract methodological prose in the six-step flow, so guidance is present and useful but not uniformly concrete or copy-paste ready, matching the 'Some concrete guidance but incomplete' anchor.

2 / 3

Workflow Clarity

The six-step process and 七项注意 are clearly sequenced and the 操作规程 enforces an explicit ordering constraint ('调查结论必须先于任何行动方案出现'), but there is no validation/verification checkpoint (e.g., how to confirm the investigation is actually sufficient) and the steps lack explicit error-recovery feedback, matching the 'Steps listed but validation gaps' anchor.

2 / 3

Progressive Disclosure

The body is well sectioned and keeps a single inline pointer ('详细原著依据见 original-texts.md'), but no bundle directories exist and that referenced file is missing, so the reference is unverified and navigation is only loosely signaled, matching the 'Some structure but could be better organized; references present but not clearly signaled' anchor.

2 / 3

Total

8

/

12

Passed

Description

82%

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 specifies what the skill does and when to trigger it, with strong natural-language trigger terms in both Chinese and English. It is somewhat held back by an abstract action vocabulary, second-person voice, and modest overlap risk with other reasoning skills.

Suggestions

Reword in third person ('Investigates firsthand before drawing conclusions') instead of 'Use this skill to...' to avoid the voice penalty.

Replace abstract verbs ('investigate first, gather firsthand facts') with more specific concrete actions (e.g., 'reads source code, runs checks, lists unknowns') to raise specificity.

Sharpen the niche so it does not read as a generic reasoning skill — name what is out of scope (e.g., not for routine implementation or execution-only tasks) to reduce conflict risk.

DimensionReasoningScore

Specificity

The description names concrete actions ('investigate first, gather firsthand facts, and let reality shape the conclusion') but these are broad abstract verbs rather than a comprehensive list of specific operations, matching the 'Names domain and some actions, but not comprehensive' anchor; the second-person 'Use this skill' phrasing further lowers specificity per the voice guideline.

2 / 3

Completeness

It explicitly answers both 'what' ('investigate first, gather firsthand facts, and let reality shape the conclusion') and 'when' ('Trigger before making claims or decisions when context is incomplete...'), satisfying the explicit-trigger anchor.

3 / 3

Trigger Term Quality

It covers natural trigger signals a user would actually voice — 'unknowns, 信息缺口, 证据不足, 领域陌生' and 'claims or decisions when context is incomplete, evidence is weak, or the domain is unfamiliar' — giving good coverage of natural terms.

3 / 3

Distinctiveness Conflict Risk

'Investigate first' is a recognizable niche but as a meta-reasoning skill it could still overlap with general analysis or planning skills, matching the 'Somewhat specific but could still overlap with similar skills' anchor.

2 / 3

Total

10

/

12

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
HughYau/qiushi-skill
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.