ml-paper-writing

Write publication-ready ML/AI/Systems papers for NeurIPS, ICML, ICLR, ACL, AAAI, COLM, OSDI, NSDI, ASPLOS, SOSP. Use when drafting papers from research repos, structuring arguments, verifying citations, or preparing camera-ready submissions. Includes LaTeX templates, reviewer guidelines, and citation verification workflows.

Quality

88%

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

77%

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

The body is highly actionable with strong, validation-gated workflows and good one-level-deep references for the reference docs. Its main weaknesses are verbosity from inline time-sensitive conference tables that duplicate the systems-conferences reference, and a missing templates/ bundle that the description promises and several workflows depend on.

Suggestions

Add the missing templates/ directory (or remove/relabel the template references) so the advertised "Includes LaTeX templates" and Workflow 4 navigation actually resolve.

Move the year-specific conference quick-reference tables into references/systems-conferences.md and keep only a compact, venue-agnostic summary inline to cut time-sensitive tokens.

Tighten or relocate the "What Reviewers Actually Read" and "Time Allocation" philosophy prose into references/writing-guide.md, since it restates context Claude largely already knows.

Dimension	Reasoning	Score
Conciseness	Mostly actionable but padded: inline year-specific conference tables ("NeurIPS 2025", "ICML 2026", "NSDI 2027") embed time-sensitive data that the rubric flags, and conference requirements are duplicated both inline and in references/systems-conferences.md. Not a 3 because these could be tightened or moved entirely to the reference file.	2 / 3
Actionability	Provides copy-paste-ready executable code throughout — bash exploration commands, a working Python `doi_to_bibtex` function, Semantic Scholar search code, LaTeX/booktabs examples, and concrete per-step checklists.	3 / 3
Workflow Clarity	Five numbered workflows (0–4) each have sequenced checklists, and the citation and template workflows include explicit validation checkpoints (verify in 2+ sources, verify the claim appears, compile template as-is before editing) with feedback/retry loops.	3 / 3
Progressive Disclosure	The six references/ files are well-signaled via a navigation table and inline links, but the body repeatedly references a templates/ directory and templates/README.md (and per-conference template dirs) that do not exist in the bundle, breaking navigation for a core advertised feature.	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.

A strong, third-person description that names concrete capabilities, lists the specific target conferences as natural triggers, and pairs a clear what with an explicit Use-when clause. It is distinguishable from generic writing skills and makes no over-claims.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — "Write publication-ready ML/AI/Systems papers", "drafting papers from research repos", "structuring arguments", "verifying citations", "preparing camera-ready submissions" — matching the multiple-specific-actions anchor.	3 / 3
Completeness	Explicitly states what it does ("Write publication-ready...") and an explicit "Use when drafting papers from research repos, structuring arguments, verifying citations, or preparing camera-ready submissions" trigger clause, satisfying both what and when.	3 / 3
Trigger Term Quality	Names the exact conference venues a user would say (NeurIPS, ICML, ICLR, ACL, AAAI, COLM, OSDI, NSDI, ASPLOS, SOSP) plus natural phrases like "drafting papers", "research repos", and "camera-ready submissions".	3 / 3
Distinctiveness Conflict Risk	Tightly scoped to publication-ready ML/AI/Systems paper writing for named venues with citation-verification workflows — a clear niche unlikely to fire for unrelated skills.	3 / 3
	Total	12 / 12 Passed

Validation

81%

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 — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (1016 lines); consider splitting into references/ and linking	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
relative_links	Relative link issues: 2 missing	Warning

	Total	13 / 16 Passed

Repository: OpenRaiser/NanoResearch
Path: skills/vendor-ai-research/ml-paper-writing/SKILL.md
Commit: 7144364

Reviewed: about 13 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.