deep-research

Generate format-controlled research reports with evidence tracking, citations, source governance, and multi-pass synthesis. This skill should be used when users request a research report, literature review, market or industry analysis, competitive landscape, policy or technical brief. Triggers: "帮我调研一下", "深度研究", "综述报告", "深入分析", "research this topic", "write a report on", "survey the literature on", "competitive analysis of", "技术选型分析", "竞品研究", "政策分析", "行业报告". V6 adds: source-type governance, AS_OF freshness checks, mandatory counter-review, and citation registry. V6.1 adds: source accessibility (circular verification forbidden, exclusive advantage encouraged).

Quality

89%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

85%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 highly actionable with a clearly sequenced, validation-rich pipeline and exemplary progressive disclosure via organized reference files. The main weakness is conciseness: the Enterprise Research workflow is restated multiple times, which could be consolidated.

Suggestions

Consolidate the Enterprise Research workflow, which currently appears as an overview (lines ~123-146), a full 'Enterprise Research Mode' section (lines ~181-260), and an 'E3-E7' recap (lines ~264-269); keep one canonical description and link to it.

Remove the duplicated 'When researching a specific company/enterprise, follow this specialized workflow...' lead-in that appears both before the Enterprise Workflow Overview and at the start of the Enterprise Research Mode section.

Tighten the source-accessibility tables and examples by moving the full 'Handling Information Black Box' worked example into a reference file, leaving only the rule and a one-line pointer inline.

Dimension	Reasoning	Score
Conciseness	Core content is lean and avoids explaining concepts Claude already knows, but the Enterprise workflow is described three times (overview, full section, and E3-E7 recap), adding notable redundant bulk.	3 / 5
Actionability	Provides copy-paste-ready status strings, registry format blocks, exact quality-gate thresholds, file paths, and a concrete black-box example covering common cases.	5 / 5
Workflow Clarity	The P0-P7 pipeline is explicitly sequenced with mandatory validation checkpoints (P3 gates, P6 counter-review finding >=3 issues, P7 spot-checks) and feedback loops for error recovery.	5 / 5
Progressive Disclosure	A dedicated Reference Files section organizes 14 verified bundle files into three groups with 'When to Load' guidance, keeping references one level deep and well-signaled.	5 / 5
	Total	18 / 20 Passed

Description

95%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 is strong: it states concrete capabilities, gives an explicit use-when clause, and lists an unusually thorough bilingual set of trigger phrases. Its only mild weakness is that the capability list could be even more granular.

Dimension	Reasoning	Score
Specificity	Names several concrete capabilities (evidence tracking, citations, source governance, multi-pass synthesis) but does not enumerate the full action set, leaving minor coverage gaps.	4 / 5
Completeness	Explicitly answers 'what' (first sentence) and 'when' ('This skill should be used when...') with concrete trigger phrases, satisfying both halves clearly.	5 / 5
Trigger Term Quality	An explicit 'Triggers:' list provides comprehensive natural phrases in both English and Chinese, covering synonyms and report-type variants users would actually say.	5 / 5
Distinctiveness Conflict Risk	Niched to research-report generation with a distinctive bilingual trigger vocabulary, giving it minimal overlap with unrelated skills.	5 / 5
	Total	19 / 20 Passed

Validation

93%

Warnings & errors only

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

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (545 lines); consider splitting into references/ and linking	Warning

	Total	15 / 16 Passed

Repository: daymade/claude-code-skills
Path: deep-research/SKILL.md
Commit: b8556f2

Reviewed: 2 days ago

Table of Contents

Discovery Implementation Validation

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.