literature-close-read

Produce a structured close-reading report from a paper's full PDF-to-Markdown text (with `## Page XX` pagination and image references) when you need to systematically extract background, research questions, methods, results, limitations, and reproducible experimental details.

Quality

68%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./scientific-skills/Evidence Insight/literature-close-read/SKILL.md

Quality

Content

46%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 skill has a clear sequenced workflow and good reference structure with real bundle files, but is noticeably verbose due to generic boilerplate sections and offers limited concrete, executable guidance — extraction detail lives in prose/checklists rather than actionable commands or code.

Suggestions

Remove the generic boilerplate sections (When Not to Use, Required Inputs, Validation and Safety Rules, Failure Handling, Input Validation, User Checkpoints) that restate guardrails Claude already follows, to improve conciseness.

Replace the vague 'pdf-extract paper.pdf > paper.md' example with a concrete, runnable conversion command (or explicitly state it is illustrative and link to a dependency doc), since it is currently the only command in the skill.

Tighten Step 5 so 'note obvious unstated limitations' is framed as 'flag only if the paper does not address them' to remove the tension with the no-speculation constraint.

Dimension	Reasoning	Score
Conciseness	The body is padded with templated, generic sections ('When Not to Use', 'Required Inputs', 'Validation and Safety Rules', 'Failure Handling', 'Input Validation', 'User Checkpoints') that restate obvious guardrails Claude already applies, inflating the file well beyond what the core workflow needs.	2 / 5
Actionability	Guidance is largely procedural prose rather than executable artifacts; the only concrete command is a vague 'pdf-extract paper.pdf > paper.md' whose options are explicitly undetermined, and the actual extraction logic is deferred to a checklist and template rather than given as concrete steps.	2 / 5
Workflow Clarity	A clear seven-step sequence with an explicit quality-check step and a 'write Not specified / never fabricate' validation discipline; minor gaps are that validation is implicit (no runnable check) and Step 5's 'note obvious unstated limitations' slightly tensions with the no-speculation rule.	4 / 5
Progressive Disclosure	The overview points cleanly to two real one-level-deep bundle files (references/guide.md and assets/deep_reading_template.md), both verified to exist and appropriately scoped; the body itself could shed inlined generic boilerplate that belongs in no separate file.	4 / 5
	Total	12 / 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.

A strong, specific description that clearly states both what it does and when to use it, with concrete extraction targets and an explicit no-inference constraint. Minor room to add common synonyms a user might say ('literature review', 'critique a paper').

Suggestions

Add one or two natural trigger synonyms like 'literature review' or 'critique a paper' to broaden the trigger_term_quality beyond the current technical phrasing.

Consider noting the file extension cue (e.g., '.md' / '.pdf') so the description fires when users reference a converted paper file by name.

Dimension	Reasoning	Score
Specificity	It names the domain and multiple concrete actions: 'extract background, research questions, methods, results, limitations, and reproducible experimental details' — comprehensive coverage of what the skill produces.	5 / 5
Completeness	Explicitly answers both 'what' (produce a structured close-reading report extracting background/methods/results/etc.) and 'when' ('when you need to systematically extract...'), with concrete trigger framing.	5 / 5
Trigger Term Quality	Strong natural terms ('paper', 'close-reading report', 'reproducible experimental details', 'PDF-to-Markdown') but lacks common synonyms like 'literature review', 'critique', or explicit file extensions the user might say.	4 / 5
Distinctiveness Conflict Risk	A clear niche (PDF-to-Markdown close reading with pagination/image handling and a no-speculation constraint) that is unlikely to fire for unrelated skills.	5 / 5
	Total	19 / 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
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
relative_links	Relative link issues: 2 suspicious	Warning

	Total	14 / 16 Passed

Repository: aipoch/medical-research-skills
Path: scientific-skills/Evidence Insight/literature-close-read/SKILL.md
Commit: 63c61d3

Reviewed: 11 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.