CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

medical-review-writer-architect

A multi-stage workflow for writing long-form medical reviews; used when the user needs to build an outline based on PubMed literature, write chapter by chapter, perform supplementary searches, and format citations; input is the review topic and project directory, output is a c...

Quality

91%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

92%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.

A well-engineered, highly actionable skill body: concrete commands with exact arguments, quantified acceptance thresholds, a state-machine-driven six-stage workflow with explicit validation and user checkpoints, and clean one-level-deep progressive disclosure into real bundle files. The only slack is minor redundancy (the sub-heading rule stated twice) and generic closing validation boilerplate.

DimensionReasoningScore

Conciseness

The body is efficient and entirely task-specific with no explanation of concepts Claude already knows, but the 2-4 sub-heading mandate is repeated nearly verbatim in Stage 1 and Stage 3, and the closing 'Input Validation'/'Quick Validation' sections read as generic template boilerplate that could be trimmed. These are minor instances of over-writing, not padded or verbose sections.

4 / 5

Actionability

Fully executable guidance throughout: exact script invocations with arguments ('python scripts/pubmed_search.py "QUERY" --section "introduction|section_1|..."', 'python scripts/check_section.py "section_1.md"', 'merge_review.py abstract.md introduction.md section_1.md ... -o Review_Final.md', 'revert_citations.py Review_Final_Formatted.md -o draft.md') plus quantified acceptance criteria (>=150 unique PMIDs, <7000 characters per chapter, no more than 2 references per sentence).

5 / 5

Workflow Clarity

Six explicitly sequenced stages with a state machine printing next-step guidance after each script run, mandatory verification after each chapter with concrete remediation loops (--post-write supplementary searches, expanding sub-headings, splitting dense citations), user checkpoints after the outline and each chapter, and a dedicated Failure Handling section. Validation and feedback loops are explicit and pervasive.

5 / 5

Progressive Disclosure

The body is an overview that links each stage to the specific stage of references/workflow-details.md or references/citation-guide.md (both files exist, are one level deep, and contain no nested references), points to assets/Review_Final.md as the format exemplar, and closes with a Resource Files section mapping the bundle. Navigation is easy and appropriately split.

5 / 5

Total

19

/

20

Passed

Description

87%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 description that clearly states a multi-stage capability set, explicit use conditions, and a distinctive PubMed-based medical-review niche. Its one real defect is that the text is cut off mid-sentence ('output is a c...'), which slightly weakens the specificity of the stated deliverable, and it would benefit from synonym coverage of review types users actually name.

Suggestions

Restore the truncated tail so the output contract is stated in full (e.g., 'output is a complete Review_Final.md with formatted PubMed citations') instead of ending at 'output is a c...'.

Add natural synonyms users would say, such as 'narrative review', 'literature review', or 'review article', alongside 'long-form medical reviews' to improve trigger coverage.

DimensionReasoningScore

Specificity

Enumerates several concrete actions ('build an outline based on PubMed literature, write chapter by chapter, perform supplementary searches, and format citations'), but the sentence is truncated mid-clause ('output is a c...'), leaving a minor gap in coverage of the deliverables. It is well above the 1-2-action anchor and just short of fully comprehensive.

4 / 5

Completeness

Explicitly answers both what ('A multi-stage workflow for writing long-form medical reviews' with enumerated actions) and when ('used when the user needs to build an outline based on PubMed literature, write chapter by chapter...') with concrete trigger phrases. Both are stated before the truncated tail, so nothing material is lost.

5 / 5

Trigger Term Quality

Natural keywords users would say are present ('medical reviews', 'PubMed', 'outline', 'chapter', 'citations'), but common synonyms such as 'literature review', 'narrative review', or 'systematic review' are missing. Coverage is good rather than comprehensive.

4 / 5

Distinctiveness Conflict Risk

Occupies a clear niche (long-form medical review writing grounded in PubMed) with distinctive triggers like 'PubMed literature' and 'chapter by chapter', making accidental triggering by unrelated skills unlikely.

5 / 5

Total

18

/

20

Passed

Validation

93%

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

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

15

/

16

Passed

Repository

aipoch/medical-research-skills

Path

scientific-skills/Academic Writing/medical-review-writer-architect/SKILL.md

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.