clinical-trial-protocol-skill

Generate clinical trial protocols for medical devices or drugs. This skill should be used when users say "Create a clinical trial protocol", "Generate protocol for [device/drug]", "Help me design a clinical study", "Research similar trials for [intervention]", or when developi...

Quality

65%

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/Protocol Design/clinical-trial-protocol-skill/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 skill is a well-architected modular workflow with strong actionability (concrete commands, exact menus, templates) and a sound progressive-disclosure structure. Its weaknesses are significant verbosity from repeated on-demand-loading/step-mapping narration and internal inconsistencies — mislabeled step headings and a non-existent referenced file — that erode workflow clarity and navigation reliability.

Suggestions

Consolidate the on-demand-loading directive and the step-to-file mapping so each appears once; remove the near-verbatim restatement of the Execution Flow in the "Implementation Requirements" section to cut substantial repetition.

Fix the mislabeled step headings: rename "**Step 2.5: Protocol Completion Menu**" and "**Step 3: Final Summary**" to reflect their true positions (post-Step-4 pause and post-Step-5 summary), and align the waypoint/filename references (e.g., the non-existent `references/05-generate-document.md` should be `references/05-concatenate-protocol.md`).

Resolve the waypoint naming inconsistencies (e.g., `02_protocol_draft.md` described as concatenated "in Step 4" vs. produced by Step 5; Final Summary citing `waypoints/protocol_complete.md` while earlier sections cite `waypoints/02_protocol_draft.md`) so the resume/checkpoint logic is unambiguous.

Dimension	Reasoning	Score
Conciseness	The body (~460 lines) repeats the on-demand-loading directive roughly four times (the EXECUTION CONTROL block, "Modular Subskill Steps", "Full Workflow Logic", and "Implementation Requirements" item 4) and restates the full step-to-file mapping twice, plus the "Implementation Requirements" section re-narrates the entire Execution Flow nearly verbatim — this is noticeably verbose with several unnecessary padded/repeated sections, landing on the score-2 anchor rather than the 3 where padding is only incidental.	2 / 5
Actionability	Quotes concrete executable guidance: `pip install -r requirements.txt` with pinned versions ("scipy >= 1.11.0", "numpy >= 1.24.0"), explicit step-to-reference-file mappings, and exact copy-paste menu text plus a full research-summary markdown template; it is the score-4 anchor (mostly executable, minor gaps) rather than 5 because some guidance is procedural narration rather than code, and the referenced requirements.txt file is not actually present in the bundle.	4 / 5
Workflow Clarity	Quotes a clear multi-step sequence (Steps 0-5) with completion-status checks, error handling ("Retry step? (Yes/No)"), resume capability, and a Protocol Completion Menu checkpoint — features that would reach 4 — but internal mislabeling (a heading literally reading "Step 2.5: Protocol Completion Menu" and another "Step 3: Final Summary" for what is actually post-Step-5) plus a duplicated step-to-file mapping with a conflicting filename leave checkpoints implicit/inconsistent, capping it at the score-3 anchor.	3 / 5
Progressive Disclosure	Quotes a well-structured modular layout with one-level-deep references in `references/` (verified present: 00-04 plus 05-concatenate-protocol.md, plus a script and an asset) and an explicit on-demand-loading philosophy — which would reach 4 — but the references tree in the body lists a file `05-generate-document.md` that does not exist (the real file is `05-concatenate-protocol.md`), a broken/contradictory reference that undermines navigation, leaving it at the score-3 anchor.	3 / 5
	Total	12 / 20 Passed

Description

81%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 clearly answers both 'what' and 'when' with an explicit trigger clause and a well-scoped niche, giving it strong distinctiveness and trigger-term quality. Its main weakness is moderate specificity (only a couple of concrete actions named) and a trailing truncation that slightly undercuts completeness.

Dimension	Reasoning	Score
Specificity	Quotes "Generate clinical trial protocols for medical devices or drugs" and "Research similar trials for [intervention]" — it names the domain and a couple concrete actions but the action list is thin relative to the full 6-step workflow (foundation/intervention/operations/document generation) described in the body, so it stops at the score-3 anchor rather than comprehensive 5.	3 / 5
Completeness	Quotes both a clear "what" ("Generate clinical trial protocols for medical devices or drugs") and an explicit "when" clause ("This skill should be used when users say..."), matching the score-4 anchor of having both with concrete triggers; it is capped below 5 because the description is truncated mid-clause ("or when developi...") so the 'when' is not perfectly explicit/complete.	4 / 5
Trigger Term Quality	Quotes five natural trigger phrases users would actually say ("Create a clinical trial protocol", "Generate protocol for [device/drug]", "Help me design a clinical study", "Research similar trials for [intervention]"), giving comprehensive natural-term coverage including synonyms and bracketed variants; it is not the 4 anchor because the phrase list is extensive rather than merely 'a few terms missing'.	5 / 5
Distinctiveness Conflict Risk	Quotes a clear niche (clinical trial protocols for medical devices or drugs) with distinct, domain-specific trigger phrases that are unlikely to fire for an unrelated skill, matching the score-5 anchor for minimal conflict risk.	5 / 5
	Total	17 / 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
skill_md_line_count	SKILL.md is long (519 lines); consider splitting into references/ and linking	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: aipoch/medical-research-skills
Path: scientific-skills/Protocol Design/clinical-trial-protocol-skill/SKILL.md
Commit: 63c61d3

Reviewed: 3 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.