grant-proposal

Draft a structured grant proposal from research ideas and literature. Supports KAKENHI (Japan), NSF (US), NSFC (China, including 面上/青年/优青/杰青/海外优青/重点), ERC (EU), DFG (Germany), SNSF (Switzerland), ARC (Australia), NWO (Netherlands), and generic formats. Use when user says "write grant", "grant proposal", "申請書", "write KAKENHI", "科研費", "基金申请", "写基金", "NSF proposal", or wants to turn research ideas into a funding application.

Quality

—

Does it follow best practices?

Impact

—

No eval scenarios have been run

Securityby

Advisory

Suggest reviewing before use

Quality

Content

77%

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

The content is highly actionable with clear sequenced workflows and validation loops, but it is verbose with redundant sections and fails progressive disclosure by keeping all agency specs and drafting guidance inline while referencing bundle files that are not present.

Suggestions

Split the per-agency Grant Type Specifications and the grant-specific drafting guidelines into a references/ file (e.g. GRANT_TYPES.md) and link to it, keeping the body as an overview.

Remove the duplicated funding-track diagram and the Parameter Pass-Through table that restates the Constants block to cut token cost.

Create the missing references/ (and scripts/) bundle directories for the referenced output-manifest.md, output-versioning.md, and extract_paper_style.py so the body's references resolve to real files.

Dimension	Reasoning	Score
Conciseness	The ~700-line body is largely justified by the multi-agency complexity, but contains redundancy: the funding/publish track is diagrammed twice (Overview and Composing sections), the Parameter Pass-Through table restates the Constants block, and 'Composing with Other Skills' overlaps the Overview — tightening would reduce tokens without losing meaning.	2 / 3
Actionability	Provides concrete, executable guidance throughout: a real shell snippet for style-ref helper resolution, a literal GRANT_STATE.json schema, drafting order, figure list, and copy-paste-ready Codex MCP prompt templates with bundle contents.	3 / 3
Workflow Clarity	Phases 0-5 are clearly sequenced with explicit 🚦/⛔ checkpoint markers and STOP instructions, plus validation feedback loops (review-revise rounds, CRITICAL/MAJOR/MINOR severity parsing) and state persistence for compaction recovery.	3 / 3
Progressive Disclosure	The body is a long monolithic wall with grant-type specs, drafting guidelines, and examples all inline; it references bundle/shared-reference files (output-versioning.md, output-manifest.md, extract_paper_style.py) but no references/, scripts/, or assets/ directories exist in the skill, so those references cannot be verified and content that should be split out is kept inline.	2 / 3
	Total	10 / 12 Passed

Description

100%

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 specific, trigger-rich, and complete, covering both the action and the use conditions across multiple languages and agencies. Its only weakness is verbosity from enumerating every agency and sub-type inline.

Suggestions

Trim the inline agency/sub-type enumeration in the description to the most common ones and rely on the body for the full list; the description is quite long for a trigger string.

Dimension	Reasoning	Score
Specificity	States the concrete action 'Draft a structured grant proposal from research ideas and literature' and enumerates multiple supported agencies (KAKENHI, NSF, NSFC, ERC, DFG, SNSF, ARC, NWO) with their sub-types — several specific concrete actions rather than vague language.	3 / 3
Completeness	Explicitly answers both 'what' ('Draft a structured grant proposal from research ideas and literature') and 'when' via the 'Use when user says ...' clause listing concrete triggers.	3 / 3
Trigger Term Quality	Includes natural terms a user would actually say across languages ('write grant', 'grant proposal', 'NSF proposal', '科研費', '写基金', '申請書', 'write KAKENHI'), giving broad coverage of common phrasings.	3 / 3
Distinctiveness Conflict Risk	Occupies a clear niche (grant/funding proposals) with distinct multilingual trigger terms unlikely to fire for unrelated skills; the agency enumeration is verbose but specific, not conflicting.	3 / 3
	Total	12 / 12 Passed

Validation

75%

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 — 12 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (696 lines); consider splitting into references/ and linking	Warning
allowed_tools_field	'allowed-tools' contains unusual tool name(s)	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
relative_links	Relative link issues: 3 suspicious	Warning

	Total	12 / 16 Passed

Repository: wanshuiyin/Auto-claude-code-research-in-sleep
Commit: fe5963c

Reviewed: about 19 hours 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.