retraction-watcher

Automatically scan reference lists and check whether cited papers have been retracted, corrected, or flagged; use before submission, review, or evidence synthesis to reduce citation risk.

Quality

64%

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

Fix and improve this skill with Tessl

tessl review fix ./scientific-skills/Evidence Insight/retraction-watcher/SKILL.md

Quality

Content

50%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 pairs genuinely useful technical detail and executable commands with a real script, but is weighed down by heavy templated boilerplate, broken/dead references, and duplicated sections that hurt token efficiency and navigation.

Suggestions

Remove the generic boilerplate sections (Risk Assessment, Security Checklist, Evaluation Criteria, Lifecycle Status, Output Contract, Failure Handling) and consolidate the overlapping Purpose/When to Use/Trigger Conditions and Error Handling/Failure Handling blocks to cut the body roughly in half.

Fix or delete dead references: the first References section and Prerequisites cite citation-formats.md, api-documentation.md, example-reports/, and requirements.txt, none of which exist — either add these files or drop the references and keep only the valid audit-reference.md link.

Integrate the py_compile/--help validation as an explicit numbered step inside the Workflow with a validate-then-proceed feedback loop, so the batch scanning process has concrete checkpoints instead of relying on a separate Quick Check section.

Dimension	Reasoning	Score
Conciseness	The 322-line body is noticeably padded with generic templated boilerplate (Risk Assessment, Security Checklist, Evaluation Criteria, Lifecycle Status, Output Contract, etc.) and redundant overlapping sections rather than over-explaining known concepts.	2 / 5
Actionability	Provides concrete, executable commands (py_compile, --help, --input manuscript.pdf --format detailed) backed by a real 732-line script, plus specific APIs and a 90% title-match threshold; minor gaps since the matching algorithm is prose rather than code.	4 / 5
Workflow Clarity	A 5-step workflow with stop-early and fallback steps exists, but for a batch operation the validation checkpoint (py_compile) is not integrated into the sequence with explicit feedback loops, capping clarity at 3.	3 / 5
Progressive Disclosure	Section structure is clear and one reference (audit-reference.md) is valid and one level deep, but the first References section and Prerequisites point to nonexistent files (citation-formats.md, api-documentation.md, example-reports/, requirements.txt) and the References section is duplicated.	3 / 5
	Total	12 / 20 Passed

Description

78%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 focused, distinct description that clearly states both capability and use context with concrete actions. The "when" guidance could be sharpened into explicit user-utterance trigger phrases to reach the top anchor.

Dimension	Reasoning	Score
Specificity	Names the domain and several concrete actions ("scan reference lists", "check whether cited papers have been retracted, corrected, or flagged"), though they are variations of one verb rather than a broad action set.	4 / 5
Completeness	Explicitly answers both "what" (scan and check retraction status) and "when" ("use before submission, review, or evidence synthesis"), but the "when" is a timing context rather than concrete user-utterance triggers.	4 / 5
Trigger Term Quality	Good coverage of natural terms ("reference lists", "cited papers", "retracted", "submission", "review", "evidence synthesis"), but missing common user utterances like "check my references" and synonym variants.	4 / 5
Distinctiveness Conflict Risk	A clear niche (retraction/correction checking of citations) with distinct triggers and minimal overlap risk against other skills.	5 / 5
	Total	17 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
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/Evidence Insight/retraction-watcher/SKILL.md
Commit: 63c61d3

Reviewed: 5 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.