research-idea

当用户提供研究资料、项目背景、实验结果、论文草稿、PR/仓库信息或自然语言线索，希望在文献调查基础上提出科学问题、凝练可证伪假设、寻找创新点或判断研究想法价值时使用。⚠️ 不适用：用户只需要完整实验方案/分析计划（优先 research-plan）、只要写文献综述正文（优先 research-literature-review）、或只要不需要文献依据的普通头脑风暴。

Quality

86%

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.

A thorough, well-structured workflow skill with clear sequencing, concrete tooling, explicit validation gates, and a clean reference layer that matches the actual bundle. The main weakness is mild repetition of failure-mode caveats across sections.

Suggestions

Consolidate the repeated insufficient/zero-candidate and hidden-workspace-path caveats into a single canonical statement referenced from each section, to reduce repetition.

Tighten the 输出 and 输出管理 sections, which overlap on default paths and hidden-workspace rules.

Dimension	Reasoning	Score
Conciseness	Dense and assumes Claude's competence without explaining research concepts, but caveats about insufficient/zero-candidate handling and hidden-workspace paths recur across sections and could be tightened.	4 / 5
Actionability	Concrete executable commands (bsk workspace init --task-root, scripts/init_workspace.py, validate_report.py, check_completion.py) and named output files are given throughout, with only minor placeholder gaps like {最终报告路径}.	4 / 5
Workflow Clarity	Phases are clearly sequenced (init → literature → candidates → novelty → review → selection → report+verify) with explicit validation checkpoints (validate_report.py, check_completion.py, Verifiers, bsk Gate, completed Gate) and feedback loops (fix-then-revalidate, re-explore once).	5 / 5
Progressive Disclosure	The body is an overview pointing to real one-level-deep references (runtime-guide.md, research-synthesis.md, report-template.md, novelty-check.md, agent-review-prompt.md, states/verifiers index files) and scripts, all of which exist in the bundle.	5 / 5
	Total	18 / 20 Passed

Description

87%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 well-constructed description: it names concrete capabilities, gives explicit trigger conditions, and proactively disambiguates against adjacent skills via an 不适用 clause. Trigger keyword coverage is strong though not exhaustive.

Dimension	Reasoning	Score
Specificity	Lists several concrete actions (提出科学问题、凝练可证伪假设、寻找创新点、判断研究想法价值) plus concrete input types (研究资料、项目背景、实验结果、论文草稿、PR/仓库信息或自然语言线索); not a fully exhaustive action list, so below 5.	4 / 5
Completeness	Explicitly states both what (propose scientific questions, falsifiable hypotheses, novelty, value judgment) and when (当用户提供…时使用) with concrete trigger phrases, plus an 不适用 exclusion list naming alternative skills.	5 / 5
Trigger Term Quality	Natural user-facing terms like 研究资料、论文草稿、PR/仓库信息、科学问题、假设、创新点 appear with synonyms, but a few common phrasings (e.g. 'brainstorm', 'novelty') are only covered indirectly via the 不适用 clause.	4 / 5
Distinctiveness Conflict Risk	Clear niche — literature-grounded research-idea generation — with explicit 不适用 routing to research-plan and research-literature-review, minimizing conflict with neighboring skills.	5 / 5
	Total	18 / 20 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
relative_links	Relative link issues: 4 deeper-than-1-level	Warning
referenced_paths_exist	Referenced path issues: 4 deeper-than-1-level	Warning

	Total	14 / 16 Passed

Repository: huangwb8/ChineseResearchLaTeX
Path: skills/research-idea/SKILL.md
Commit: dd19421

Reviewed: 4 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.