CtrlK
BlogDocsLog inGet started
Tessl Logo

literature-review

Find, verify, and synthesize scientific literature — from "what's the seminal paper for X" through full multi-source reviews. Covers grounding claims in real retrieved sources, avoiding fabricated citations, handling retractions, and calibrating confidence to evidence strength.

65

Quality

80%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./configs/microservice/bff-service/configs/agent-skills/claude-science/literature-review/SKILL.md
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 highly actionable, well-sequenced instruction skill with concrete helpers, an explicit citation format, and validation checkpoints baked into the workflow. Its main weakness is verbosity — the argumentative prose earns its point once but repeats — and the absence of any bundle split for material that could live in reference files.

Suggestions

Tighten the rhetorical/argumentative passages (e.g., the 'bibliography in paragraph costume' and 'memory supplies plausible names' paragraphs) to one lean statement each; the point lands once and frees token budget.

Move the detailed citation-format and style_pass rules into a short reference file (e.g., references/citation-format.md) and keep SKILL.md a lean overview, improving progressive disclosure.

Note explicitly that kernel.py lives alongside SKILL.md (the exec path uses a placeholder) so the helper contract is unambiguous even when the bundle is inspected in isolation.

DimensionReasoningScore

Conciseness

Mostly efficient and avoids explaining basics Claude knows, but several paragraphs are padded with rhetorical flourish ('a bibliography in paragraph costume', 'memory supplies plausible names, not the paper's') and argumentative prose that could be tightened into leaner instruction.

2 / 3

Actionability

Provides concrete, executable guidance — named helpers (verify_dois, crossref_lookup, search_openalex, expand_citations, style_pass), specific env vars, an exact inline citation format with a URL-encoding rule, and a concrete scale ('on the order of fifteen or more distinct primary-paper DOIs').

3 / 3

Workflow Clarity

Clear sequenced workflow — exec kernel.py, read the request type, run the sweep, walk the citation graph backward/forward, check retractions, synthesize, run style_pass once, save — with explicit validation checkpoints (DOI resolution 'checkable in five seconds', retraction checks, style_pass as a lint-not-gate with 'do not loop').

3 / 3

Progressive Disclosure

The skill ships as a single dense SKILL.md with no bundle files (references/scripts/assets absent); sections are well-organized but substantial inline content (style_pass rules, citation-format details, env-var configuration) could be split out for easier navigation.

2 / 3

Total

10

/

12

Passed

Description

82%

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 specific, distinctive description that names concrete actions and natural trigger phrasings, but it omits an explicit 'Use when...' clause so the 'when to invoke' guidance is only implied. Adding a sentence naming the request types that should trigger it would lift completeness to full marks.

Suggestions

Add an explicit 'Use when...' clause naming the request types that should trigger this skill (e.g., 'Use when the user asks for a seminal paper, an evidence synthesis, a methods comparison, or where the gaps are in a field').

Surface more common lay phrasings (e.g., 'find me a paper on', 'what does the research say about') to broaden trigger coverage beyond the domain-expert register.

DimensionReasoningScore

Specificity

Lists multiple concrete actions — 'Find, verify, and synthesize', 'grounding claims in real retrieved sources', 'avoiding fabricated citations', 'handling retractions', 'calibrating confidence to evidence strength' — matching the anchor for multiple specific concrete actions.

3 / 3

Completeness

Clearly answers 'what' (find/verify/synthesize, grounding, retractions, confidence calibration) but has no explicit 'Use when...' trigger clause, so 'when' is only implied — capped at 2 per the guidelines.

2 / 3

Trigger Term Quality

Includes natural user phrasings a domain expert would say — 'what's the seminal paper for X', 'scientific literature', 'citations', 'reviews' — giving good coverage of natural terms.

3 / 3

Distinctiveness Conflict Risk

Occupies a clear niche — scientific literature retrieval, citation grounding, retraction handling — with distinct triggers unlikely to overlap with unrelated skills.

3 / 3

Total

11

/

12

Passed

Validation

87%

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

Validation14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

metadata_version

'metadata.version' is missing

Warning

metadata_field

'metadata' should map string keys to string values

Warning

Total

14

/

16

Passed

Repository
UnicomAI/wanwu
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.