aris-research-refine

Turn a vague research direction into a problem-anchored, elegant, frontier-aware, implementation-oriented method plan via iterative GPT-5.4 review. Use when the user says "refine my approach", "帮我细化方案", "decompose this problem", "打磨idea", "refine research plan", "细化研究方案", or wants a concrete research method that stays simple, focused, and top-venue ready instead of a vague or overbuilt idea.

Quality

83%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Critical

Do not install without reviewing

Quality

Content

77%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 a well-sequenced, highly actionable multi-phase workflow with strong feedback loops and checkpoint recovery. Its main weaknesses are verbosity from repeated principles and overlapping report templates, and a monolithic inline structure with no progressive disclosure to bundle files.

Suggestions

Split the large proposal template, the reviewer prompt blocks, and the report templates into reference files under references/ and link to them one level deep, reducing the inline SKILL.md to an overview plus workflow.

Deduplicate the 'Key Rules' section against the four principles in 'Overview' — keep one canonical statement of each principle and remove the restatement.

Merge the overlapping round-by-round tables in REVIEW_SUMMARY.md and REFINEMENT_REPORT.md into a single shared score-evolution/round-log structure referenced by both reports.

Dimension	Reasoning	Score
Conciseness	Mostly efficient for a complex multi-phase workflow, but ~730 lines with noticeable padding: the four governing principles are restated in 'Key Rules', and REVIEW_SUMMARY / REFINEMENT_REPORT templates overlap heavily (both carry round-by-round tables and score evolution), so it could be tightened — fitting anchor 3 rather than 4.	3 / 5
Actionability	Provides copy-paste-ready Codex MCP call blocks with full reviewer prompts, a concrete REFINE_STATE.json schema, exact file paths, and complete proposal/report templates — fully executable guidance for an orchestration skill.	5 / 5
Workflow Clarity	Phases 0–5 are clearly sequenced with an explicit checkpoint after every phase, a stop condition (score >= 9 / READY / no drift), MAX_ROUNDS cap, and a review→revise→re-review feedback loop with checkpoint recovery — matching the anchor-5 example with validation and feedback loops.	5 / 5
Progressive Disclosure	No bundle files exist (references/, scripts/, assets/ absent) and all content — the large proposal template, reviewer prompts, and multiple report templates — is inlined in a single 730-line SKILL.md; section headers give structure, but content that could be split into separate reference files is inline with no one-level-deep references, fitting anchor 3.	3 / 5
	Total	16 / 20 Passed

Description

90%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 clearly states what the skill does and when to use it, with comprehensive bilingual trigger phrases and a distinct niche. Its only weakness is specificity — the action is described as one adjective-heavy composite rather than a list of discrete concrete actions.

Dimension	Reasoning	Score
Specificity	Names the domain (research method refinement) and one composite action ('Turn a vague research direction into a ... method plan via iterative GPT-5.4 review'), but does not list multiple discrete concrete actions; the qualifiers 'problem-anchored, elegant, frontier-aware, implementation-oriented' are adjective-loaded rather than separate actions, fitting anchor 3 better than 4.	3 / 5
Completeness	Explicitly answers both what ('Turn a vague research direction into a ... method plan via iterative GPT-5.4 review') and when ('Use when the user says ...') with concrete trigger phrases, matching the anchor-5 example.	5 / 5
Trigger Term Quality	Comprehensive natural trigger coverage including synonyms in two languages ('refine my approach', 'decompose this problem', 'refine research plan', '帮我细化方案', '打磨idea', '细化研究方案') — phrases a user would naturally say, matching the comprehensive-synonyms anchor.	5 / 5
Distinctiveness Conflict Risk	Occupies a clear niche (turning vague ideas into refined method plans) with distinct refinement-oriented triggers that do not overlap with sibling skills like idea-creation or experiment-planning; minimal 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
skill_md_line_count	SKILL.md is long (740 lines); consider splitting into references/ and linking	Warning
allowed_tools_field	'allowed-tools' contains unusual tool name(s)	Warning

	Total	14 / 16 Passed

Repository: OpenLAIR/dr-claw
Path: skills/aris-research-refine/SKILL.md
Commit: 4b35756

Reviewed: 5 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.