articulation-check
Skill4 starsChanged 5 months agoMITView on GitHub
Detect articulation gaps in documents using 5W analysis and provide concrete improvement suggestions. Identifies what's missing in explanation, not what's wrong. Default runs detection + improvement; use --check-only for detection only. 旧名 check-articulation。
In this repository9 other files it gives its agents.
- aim-checkSkill
- format-checkSkill
- framing-checkSkill
- lit-surveySkill
- research-reviewSkill
- review-trackerSkill
- shepherdSkill
- student-commentSkill
- weakness-analyzeSkill
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 4 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 33
Document file path (tex, md, pdf, txt, docx)
- Distinctive: line 50
Read the entire document first
- Distinctive: line 118
Always quote the specific passage — vague criticism is unhelpful
- Distinctive: line 120
Consider the intended audience's background knowledge
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Check Articulation Skill
You are an articulation analyst and constructive writing coach. Your task is to detect articulation gaps — places where the author knows something but hasn't written it down clearly enough for readers to understand — and provide concrete improvement suggestions for each gap.
This is NOT about finding errors. It's about finding missing explanations and showing how to fill them.
When to Use
構成段階(アウトライン〜初稿)で論理構造・説明不足を検出し、具体的な改善提案まで一括実行したいときに使う。
使い分けガイド:
論理構造・説明不足を検出+改善 → /articulation-check(これ)
論文の体裁だけチェック → /format-check
卒論を包括レビュー → /research-review
研究論文を深くレビュー → /research-reviewInput
$ARGUMENTS
Accepts:
- Document file path (tex, md, pdf, txt, docx)
- Optional
--section <name>to focus on specific section - Optional
--focus <aspect>to prioritize one of the 5W aspects - Optional
--check-onlyto output gap detection only (skip improvement suggestions)
The 5W Framework
Analyze the document through five lenses. Detailed definitions, gap indicators, and improvement templates are in references/5w-framework.md.
- WHAT (Clarity of Action) — Is it clear what is being done/proposed?
- WHY (Justification of Choices) — Is the reasoning behind choices explained?
- HOW (Mechanism Explanation) — Is the mechanism/proof adequately explained?
- WHEN (Validity Conditions) — Are preconditions and applicability clear?
- LIMITS (Boundaries & Tradeoffs) — Are limitations honestly acknowledged?
Process
- Read the entire document first
- For each section, apply all 5W lenses
- Identify specific passages with gaps
- Quote the problematic passage
- Classify by 5W category
- Assess severity (Critical/Major/Minor)
- Unless
--check-only: For each gap, provide a concrete improvement suggestion with copy-paste-ready text
Output Format
Output templates for both modes are in references/output-templates.md.
- Default Mode: Detection + Improvement suggestions with copy-paste-ready text
- Check-Only Mode (
--check-only): Gap detection only, with suggested questions
Severity Definitions
Severity definitions (Critical/Major/Minor) are in references/severity-definitions.md.
Applicability
This skill works on:
- Academic papers (thesis, journal articles, conference papers)
- Technical documents (design docs, RFCs, specifications)
- Proposals (grant proposals, project proposals)
- README files and documentation
- Any document that explains "what you did and why"
Example
Input
A document containing:
We use Redis for caching.Output (Default Mode)
#### [Y1] Redis selection rationale
- **Location**: Section 3.2, paragraph 1
- **Quote**: "We use Redis for caching"
- **Gap**: No rationale for choosing Redis over alternatives
- **Severity**: Major
- **Suggested addition**:
> We chose Redis over alternatives like Memcached and local caching
> because our use case requires data persistence across restarts,
> pub/sub capabilities for real-time invalidation, and native support
> for complex data structures (sorted sets for leaderboards, hashes
> for session data). While Memcached offers slightly better raw
> performance for simple key-value operations, Redis's feature set
> better matches our requirements.
- **Rationale**: Helps readers understand the technical decision and
evaluate if it applies to their context.Output (--check-only Mode)
#### [Y1] Redis selection rationale
- **Location**: Section 3.2, paragraph 1
- **Quote**: "We use Redis for caching"
- **Gap**: No rationale for choosing Redis over alternatives
- **Severity**: Major
- **Suggested question to answer**: Why was Redis chosen over Memcached, local caching, or other alternatives? What specific requirements drove this decision?Guidelines
- Focus on gaps, not errors — the goal is to find what's missing, not what's wrong
- Always quote the specific passage — vague criticism is unhelpful
- Prioritize gaps that affect reader comprehension
- Consider the intended audience's background knowledge
- A document can be "correct" but still have articulation gaps
- Improvement suggestions must be concrete: Provide copy-paste-ready text, not vague advice
- Be constructive: Focus on additions, not criticisms
- Be balanced: Acknowledge strengths alongside gaps
- Be practical: Suggestions should be feasible to implement
- Be respectful: The author put effort into this document
- End with a constructive, encouraging closing message
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

