paper-search

Search, download, and read academic papers from 20+ sources (arXiv, PubMed, Semantic Scholar, CrossRef, etc). Use when the user asks to find papers, search for research, look up academic literature, download a paper PDF, or extract text from a paper.

Quality

89%

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

86%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 an efficient, executable CLI reference with a clear end-to-end workflow and clean organization appropriate to a single-file skill. The main improvement opportunities are showing concrete paper_id formats per source and adding a brief failure-handling note for search/download steps.

Suggestions

Add one example showing what a <paper_id> looks like for a couple of sources (e.g., arXiv vs. PubMed identifiers) so download/read commands can be constructed without guessing.

Include a short validation note in the workflow, e.g., check the download command's reported path exists and retry with a different source if a search returns no results.

Show one sample invocation end-to-end (e.g., search on arxiv then read that paper_id) to anchor the abstract command templates.

Dimension	Reasoning	Score
Conciseness	The body is a lean CLI reference: every line is either an executable command, a flag definition, or an operational note ("Config warnings go to stderr and can be ignored"), with zero explanation of concepts Claude already knows. It matches the 5 anchor ("every token earns its place") and is clearly above the 4 anchor, which allows minor trimmable over-explanation.	5 / 5
Actionability	Commands are concrete and executable with flags and defaults documented ("paper-search search \"<query>\" -n <max_per_source> -s <sources> -y <year>"), including install instructions. Not a 5 because the <paper_id> format is never exemplified (e.g., what an arXiv vs. PubMed identifier looks like), leaving a minor gap when constructing download/read commands; it is well above the 3 anchor since nothing is pseudocode.	4 / 5
Workflow Clarity	The four-step workflow (search with targeted sources, present a table, read for full text, download for PDF) is a clear, correctly sequenced procedure with per-step expected outputs. Not a 5 because there are no validation checkpoints (e.g., confirm the download succeeded or handle a source returning no results); not a 3 since the sequence is complete and explicit rather than having gaps — and no destructive or batch cap applies.	4 / 5
Progressive Disclosure	This is a single-file skill with no references/, scripts/, or assets/ directories; at under 50 lines with well-organized sections (CLI Usage, Output, Sources, Workflow) it fully fits the simple-skill exception where well-organized sections alone warrant a 5. No content exists that should be split out, and no references are buried.	5 / 5
	Total	18 / 20 Passed

Description

92%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.

This is a strong description: it clearly states what the skill does and when to use it, with concrete natural-language triggers and a distinct academic-literature niche. The only minor gap is that a few common synonym phrases (e.g., "literature review", "citations") are not covered.

Dimension	Reasoning	Score
Specificity	The description lists three concrete actions ("Search, download, and read academic papers") and enumerates named sources ("arXiv, PubMed, Semantic Scholar, CrossRef"), giving comprehensive coverage of the skill's capabilities in third person. It exceeds the 4 anchor ("several specific actions; minor gaps") because all core capabilities of the skill are explicitly named.	5 / 5
Completeness	It explicitly answers both what ("Search, download, and read academic papers from 20+ sources") and when ("Use when the user asks to find papers, search for research..."), with concrete trigger phrases. This matches the 5 anchor exactly; the when-clause is explicit rather than weakly implied.	5 / 5
Trigger Term Quality	Phrases like "find papers", "search for research", "look up academic literature", "download a paper PDF", and "extract text from a paper" are natural user utterances with good coverage. Not a 5 because common synonyms such as "literature review", "citations", or "papers on <topic>" are absent, though more than just a few terms are covered, placing it above the 3 anchor.	4 / 5
Distinctiveness Conflict Risk	The academic-paper niche with named sources (arXiv, PubMed, Semantic Scholar) and distinctive triggers ("download a paper PDF", "extract text from a paper") gives minimal conflict risk with other skills. It fits the 5 anchor's "clear niche with distinct triggers".	5 / 5
	Total	19 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: openags/paper-search-mcp
Path: claude-code/SKILL.md
Commit: 808e462

Reviewed: 1 hour 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.