research-idea

当用户提供任意研究资料、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“提出科学问题”“凝练研究假设”“找创新点”“判断一个想法是否值得做”时使用。该 skill 通过 research-topic-extractor 与 research-literature-review 做查新，并用 parallel-vibe 默认 3 轮串行独立审查打磨多个“科学问题-可证伪假设”候选，最终输出 Research-Idea_{github仓库名}_{pr名}_{时间戳}.md。⚠️ 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）、或只要普通头脑风暴且不需要查新。

Quality

78%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills/research-idea/SKILL.md

Quality

Content

70%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 well-sequenced, actionable multi-step workflow with explicit validation and feedback loops, weakened mainly by repeated install-path triples and by inlining content that already has dedicated reference files the body fails to point to.

Suggestions

Reference the existing bundle files from the body: point step 3 to references/novelty-check.md and step 4 to references/agent-review-prompt.md instead of inlining their criteria, so progressive disclosure is one level deep and clearly signaled.

Collapse the three repeated install-root variants (./, ~/.codex/skills/, ~/.claude/skills/) for init and validate into a single example with a one-line note about alternate install locations to cut token cost.

Replace the placeholder {第 ${round} 轮审查指令} in the parallel-vibe loop with the concrete review prompt (or an explicit reference to agent-review-prompt.md) so the command is copy-paste ready.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient and assumes Claude's competence, but repeats the workspace path template and triples each script invocation across three install roots (./, ~/.codex/skills/, ~/.claude/skills/) for both init and validate, and the 质量要求 section reiterates earlier constraints, so it could be tightened.	3 / 5
Actionability	It provides executable bash (init_workspace.py, the parallel-vibe round loop, validate_report.py) plus concrete JSON field schemas (theme.json, novelty-decision.json) and a report-template reference; minor gaps come from placeholder prompts like {第 ${round} 轮审查指令} and reliance on external skills whose interfaces are only partially specified.	4 / 5
Workflow Clarity	The six-step main flow is clearly sequenced with an explicit validation feedback loop in step 6 (run validate_report.py, 若校验失败先修复报告再交付), an early-fail dependency check at init, and a novelty re-loop (all 已充分研究 → return to step 2), so checkpoints are explicit.	5 / 5
Progressive Disclosure	Bundle references exist (references/agent-review-prompt.md, novelty-check.md, report-template.md and three scripts) but the body inlines the novelty criteria and review-prompt content instead of linking to those dedicated files, and references are plain-text mentions rather than signaled links, so structure is present but not clearly navigated.	3 / 5
	Total	15 / 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 strong, third-person description that clearly states what the skill does, when to trigger it, and where its boundaries lie relative to sibling skills. The main weakness is mild: the capability framing is process-oriented rather than a maximally explicit action list.

Dimension	Reasoning	Score
Specificity	The description lists several concrete actions — 查新 via research-topic-extractor/research-literature-review, parallel-vibe 默认 3 轮串行独立审查打磨候选, and a named output file Research-Idea_{...}.md — but the capability list leans more on process than a comprehensive enumeration of discrete actions, leaving minor coverage gaps.	4 / 5
Completeness	It explicitly answers both what (查新 + 多轮独立审查打磨多个科学问题-可证伪假设候选并输出报告) and when (当用户提供...希望...时使用) with concrete trigger phrases, and adds an explicit 不适用 boundary section.	5 / 5
Trigger Term Quality	It surfaces natural user phrases like 提出科学问题, 凝练研究假设, 找创新点, 判断一个想法是否值得做 alongside input types (研究资料, 论文草稿, PR/仓库信息); good coverage with synonyms, though a few common phrasings are absent.	4 / 5
Distinctiveness Conflict Risk	A clear niche (科学问题-可证伪假设 candidates) with an explicit 不适用 list redirecting research-plan, research-literature-review, and plain brainstorming to other skills, minimizing conflict risk.	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
metadata_version	'metadata.version' is missing	Warning
metadata_field	'metadata' should map string keys to string values	Warning

	Total	14 / 16 Passed

Repository: huangwb8/ChineseResearchLaTeX
Path: skills/research-idea/SKILL.md
Commit: f9a7f2c

Reviewed: 3 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.