CtrlK
BlogDocsLog inGet started
Tessl Logo

ml-paper-writing

Write publication-ready ML/AI papers for NeurIPS, ICML, ICLR, ACL, AAAI, COLM. Use when drafting papers from research repos, conducting literature reviews, finding related work, verifying citations, or preparing camera-ready submissions. Includes LaTeX templates, citation verification workflows, and paper discovery/evaluation criteria.

70

Quality

88%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Critical

Do not install without reviewing

SKILL.md
Quality
Evals
Security

Quality

Content

77%

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

A thorough, actionable skill with strong workflow structure, but it is over-long for the body: redundant sections and inline philosophy that already has a reference file inflate the token budget, and the missing templates/ bundle leaves dangling references.

Suggestions

Merge the two citation sections into the single canonical 'references/citation-workflow.md' reference and keep only a short pointer plus the golden rule in the body to remove the duplication.

Consolidate the three proactivity/draft-first passages ('Core Philosophy', 'Balancing Proactivity and Collaboration', Workflow 0 Step 5) into one concise section.

Either ship the templates/ bundle or remove the templates/ links and the 'LaTeX Templates' quick-reference tables that depend on them, so no referenced path is broken.

DimensionReasoningScore

Conciseness

Mostly high-signal domain guidance, but the body is verbose at ~1100 lines: the citation-hallucination rule is restated in two full sections ('Never Hallucinate Citations' and 'Citation Workflow (Hallucination Prevention)'), and the 'be proactive / draft first' theme recurs across 'Core Philosophy', 'Balancing Proactivity and Collaboration', and Workflow 0 Step 5. It also explains reviewer behavior and double-blind review that Claude largely already knows.

2 / 3

Actionability

Provides copy-paste-ready bash (find/grep/latexmk), LaTeX snippets, arXiv URL patterns, booktabs table templates, and concrete rules like the 5-sentence abstract formula and the do/don't citation matrix. Guidance is specific and executable.

3 / 3

Workflow Clarity

Multiple numbered workflows (0-5) with checkbox progress trackers, and the citation workflow has an explicit verify-or-mark-placeholder feedback loop ('If any step fails -> mark as placeholder and report it explicitly').

3 / 3

Progressive Disclosure

The references/ files are real and clearly signaled via tables and links one level deep, but the body repeatedly points to a templates/ directory (templates/neurips2025/, templates/README.md) that does not exist in the bundle, and large chunks of writing philosophy are inlined despite a dedicated references/writing-guide.md.

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, complete, and distinct, written in clean third-person imperative voice. It cleanly satisfies the highest anchor on every dimension.

DimensionReasoningScore

Specificity

Lists multiple concrete actions such as 'drafting papers from research repos', 'conducting literature reviews', 'finding related work', 'verifying citations', and 'preparing camera-ready submissions', matching the level-3 anchor of multiple specific concrete actions.

3 / 3

Completeness

Clearly answers both 'what' (write publication-ready ML/AI papers) and 'when' via an explicit 'Use when drafting papers...conducting literature reviews...verifying citations...' clause, satisfying the level-3 anchor.

3 / 3

Trigger Term Quality

Uses natural phrases a user would say ('drafting papers', 'literature reviews', 'related work', 'citations', 'camera-ready submissions') plus the concrete venue keywords (NeurIPS, ICML, ICLR, ACL, AAAI, COLM), giving good coverage of natural terms.

3 / 3

Distinctiveness Conflict Risk

Occupies a clear niche (publication-ready ML/AI papers for named top conferences) with distinct triggers that would not plausibly fire for unrelated skills.

3 / 3

Total

12

/

12

Passed

Validation

75%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation12 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

skill_md_line_count

SKILL.md is long (1123 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, 2 deeper-than-1-level

Warning

referenced_paths_exist

Referenced path issues: 7 deeper-than-1-level

Warning

Total

12

/

16

Passed

Repository
Galaxy-Dawn/claude-scholar
Reviewed

Table of Contents

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.