indication-dossier

Build a source-backed biomedical indication dossier. Use when a research task asks for disease biology, target rationale, patient segmentation, biomarkers, trials, drugs, competitive landscape, or translational evidence.

Quality

84%

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

Quality

Content

78%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 content is a lean, well-organized instruction-only skill with a clear sequenced workflow and an embedded quality-validation step. Its main gap is actionability: the steps describe categories of action rather than executable specifics or an output format, which leaves Claude to infer how to structure the saved dossier and provenance sidecar.

Suggestions

Add a brief output template or schema for the dossier and provenance sidecar (e.g., expected headings and a JSON sidecar shape) so the Save step is concrete rather than implied.

Tighten step 1 by naming the exact fields to capture (indication, segment, intervention, decision) instead of a single imperative, mirroring the specificity the description already provides.

Add an explicit validate-then-fix loop (e.g., re-check flagged confounders and re-rank claims before saving) to push workflow clarity from 4 to 5.

Dimension	Reasoning	Score
Conciseness	The ~12-line body is lean and efficient, names concrete source catalogs without explaining what they are, and assumes Claude's competence; every token earns its place, matching anchor 5.	5 / 5
Actionability	It gives concrete source categories and an evidence-separation structure, but the steps are high-level imperatives ("Define", "Gather", "Save") with no executable commands, output schema, or format spec for the provenance sidecar, matching anchor 3 (some concrete guidance but incomplete).	3 / 5
Workflow Clarity	The 5-step sequence is clearly ordered with a validation checkpoint in step 4 (rank by source strength, flag missing cohorts/confounders) plus a safety guardrail; it does not reach 5 because there is no explicit validate-fix-retry feedback loop for the save step.	4 / 5
Progressive Disclosure	No bundle files exist (references/scripts/assets all absent) and the skill is under 50 lines with a well-organized heading + numbered workflow + guardrail structure, so the simple-skill exception applies and it scores 5.	5 / 5
	Total	17 / 20 Passed

Description

90%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: third-person, explicit Use-when trigger clause, comprehensive natural keyword coverage, and a clearly distinct biomedical niche. Its only weakness is that the "what" states a single composite action rather than enumerating several specific actions, capping specificity at 3.

Dimension	Reasoning	Score
Specificity	"Build a source-backed biomedical indication dossier" names the domain and one concrete deliverable action, but does not enumerate multiple specific actions; it falls at anchor 3 (names domain and 1-2 concrete actions) rather than 4 because coverage is not comprehensive.	3 / 5
Completeness	It explicitly answers both "what" (Build a source-backed biomedical indication dossier) and "when" ("Use when a research task asks for...") with concrete trigger phrases, matching anchor 5; not below because the when-clause is fully explicit.	5 / 5
Trigger Term Quality	"disease biology, target rationale, patient segmentation, biomarkers, trials, drugs, competitive landscape, or translational evidence" provides comprehensive, natural keyword coverage including synonyms a researcher would actually say, matching anchor 5.	5 / 5
Distinctiveness Conflict Risk	The biomedical indication-dossier niche with domain-specific triggers (target rationale, patient segmentation, biomarkers) is clearly distinct from generic skills with minimal conflict risk, matching anchor 5.	5 / 5
	Total	18 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: companion-inc/feynman
Path: skills/indication-dossier/SKILL.md
Commit: 95c51d9

Reviewed: about 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.