irb-application-assistant

Assists researchers with Institutional Review Board (IRB) application tasks, including drafting informed consent documents, reviewing research protocols for compliance, generating application forms, and preparing submission checklists. Use when the user mentions IRB, Instituti...

Quality

67%

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/Academic Writing/irb-application-assistant/SKILL.md

Quality

Content

62%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's content shines on workflow clarity — a well-sequenced IRB submission workflow with real validation checkpoints and feedback loops — and offers concrete executable commands, but it is noticeably verbose with repeated description text and generic boilerplate, and its progressive-disclosure references all point to bundle files that do not exist.

Suggestions

Remove the verbatim description pasted twice (in 'When to Use' and 'Key Features') and collapse the overlapping py_compile/--help sections (Quick Check, Audit-Ready Commands, Example Usage) into one to cut redundancy.

Delete or compress the generic process boilerplate (Output Requirements, Error Handling, Input Validation, Response Template) that re-states concepts Claude already knows, keeping only IRB-specific guidance.

Create the referenced bundle files (references/guide.md, references/audit-reference.md, references/examples/, scripts/main.py) so the signaled one-level-deep references actually resolve, or remove the broken links.

Dimension	Reasoning	Score
Conciseness	The body pastes the full description verbatim twice ('When to Use' and 'Key Features'), repeats the same py_compile/--help commands across four overlapping sections (Example Usage, Quick Check, Audit-Ready Commands, Quick Start), and carries generic process boilerplate (Output Requirements, Error Handling, Input Validation, Response Template) that states concepts Claude already knows — matching 'noticeably verbose; several unnecessary explanations or padded sections'.	2 / 5
Actionability	It provides concrete, copy-pasteable commands with real flags for each capability — 'python scripts/main.py --task consent --protocol protocol.json --population ... --risk-level minimal --output consent_form.docx' — covering the common IRB cases, with only minor gaps (illustrative filenames and a script that is not bundled), fitting the 'mostly executable guidance' anchor.	4 / 5
Workflow Clarity	The 'Recommended Workflow' is a clear six-step sequence with explicit validation checkpoints and feedback loops ('review errors in validation_report.txt → fix each issue → re-run --task validate-submission → only proceed when the report shows zero blocking errors'), plus a Quality Checklist, matching the 'clear sequence with explicit validation steps; feedback loops for error recovery; checklists' anchor.	5 / 5
Progressive Disclosure	References are clearly signaled via markdown links ('references/guide.md', 'references/audit-reference.md', 'references/examples/') but none of these files (nor scripts/main.py) exist in the bundle, and substantial content is inlined that could live in separate files, fitting 'some structure but could be better organized; references present but not clearly signaled'.	3 / 5
	Total	14 / 20 Passed

Description

72%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.

The description is strong on specificity and distinctiveness, naming four concrete IRB actions in a clear niche, but its trigger and completeness scores are held back because the 'Use when...' clause is literally truncated mid-word ('Instituti...'), cutting off the full set of natural trigger terms.

Suggestions

Complete the truncated trigger clause so it reads the full 'Institutional Review Board' and includes the natural trigger phrases (research ethics, human subjects research, protocol review, informed consent) rather than ending at 'Instituti...'.

Add the natural synonyms and variations a researcher would actually say — 'human subjects research', 'research ethics review', 'IRB submission' — to broaden trigger-term coverage beyond the single keyword 'IRB'.

Dimension	Reasoning	Score
Specificity	Lists four concrete, distinct actions — 'drafting informed consent documents', 'reviewing research protocols for compliance', 'generating application forms', and 'preparing submission checklists' — giving comprehensive coverage of the IRB domain, matching the score-5 anchor.	5 / 5
Completeness	It clearly answers 'what' (four concrete IRB actions) and includes a 'Use when...' trigger clause, but the 'when' is truncated mid-word and therefore not fully explicit, landing at 'both what and when; when could be more explicit' rather than a clean 5.	4 / 5
Trigger Term Quality	The 'Use when' clause names 'IRB' and begins 'Instituti...' but is truncated mid-word, so synonyms users naturally say (research ethics, human subjects research, protocol review, informed consent) are cut off — only one fully-formed natural keyword is present, fitting the 'some relevant keywords but missing common variations' anchor.	3 / 5
Distinctiveness Conflict Risk	IRB application assistance is a clear niche with distinct triggers and low conflict risk, though it could overlap slightly with general academic-writing or research-compliance skills, fitting the 'mostly distinct; minor overlap risk' anchor.	4 / 5
	Total	16 / 20 Passed

Validation

81%

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 — 13 / 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: 1 missing	Warning
referenced_paths_exist	Referenced path issues: 20 missing	Warning

	Total	13 / 16 Passed

Repository: aipoch/medical-research-skills
Path: scientific-skills/Academic Writing/irb-application-assistant/SKILL.md
Commit: 63c61d3

Reviewed: 9 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.