adverse-event-narrative

Generates CIOMS I-compliant ICSR narratives from adverse event case data for FDA and EMA regulatory submission. Includes temporal analysis, MedDRA coding, causality assessment using WHO-UMC or Naranjo criteria, and multi-format output.

Quality

—

Does it follow best practices?

Impact

—

No eval scenarios have been run

Securityby

Passed

No known issues

Quality

Content

55%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body has a strong, well-sequenced workflow with validation checkpoints, but it is somewhat verbose and its progressive disclosure is broken: referenced files and imported modules do not exist, while the actual bundled reference and script files are never surfaced.

Suggestions

Replace the fabricated References list with the real bundle files (CIOMS_I_Guidelines.md, ICSR_Template.md, MedDRA_Reference.md, Quick_Reference.md, sample_case_001.json, sample_case_minimal.json) and link them with clear one-level-deep pointers.

Fix the code examples to use only the bundled `scripts/main.py` CLI, or remove imports from non-existent modules (narrative_generator, temporal_analyzer, causality_assessor) so snippets are executable.

Trim redundancy — de-duplicate the 'Audit-Ready Commands' entries and consolidate the overlapping 'Overview'/'Key Capabilities' and 'Core Capabilities' sections.

Dimension	Reasoning	Score
Conciseness	Mostly efficient domain content (CIOMS sections, WHO-UMC categories, checklists) but includes redundancy — the 'Audit-Ready Commands' duplicates `python -m py_compile scripts/main.py` and repeats `python scripts/main.py --help`, and 'Overview'/'Key Capabilities' overlap with 'Core Capabilities' — so it 'could be tightened'.	2 / 3
Actionability	Provides concrete code examples and CLI commands, but the Python snippets import from `scripts.narrative_generator` (NarrativeGenerator, analyze_timeline, assess_causality) whose module does not exist — only `main.py` is bundled — so the code is not actually executable/copy-paste ready, fitting 'Some concrete guidance but incomplete; missing key details'.	2 / 3
Workflow Clarity	The 8-step Workflow has an explicit sequence with validation (step 2), two Checkpoint steps (3 and 6), and a Fallback (step 8), plus a pre/post-generation Quality Checklist — a clear sequence with explicit validation and feedback loops, matching the top anchor.	3 / 3
Progressive Disclosure	The References section lists 7 files (e.g. `cioms_i_guidelines.pdf`, `meddra_coding_guide.md`, `who_umc_causality.md`) none of which exist, while the actual bundle (CIOMS_I_Guidelines.md, ICSR_Template.md, MedDRA_Reference.md, Quick_Reference.md, sample_case JSONs) is unreferenced; the Scripts section likewise lists 8 files but only main.py exists — navigation is broken, fitting 'poor organization' rather than well-signaled one-level-deep references.	1 / 3
	Total	8 / 12 Passed

Description

67%

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 specific and occupies a clear, distinctive regulatory niche, but it omits an explicit 'Use when...' trigger clause and lacks conversational trigger variations, capping completeness and trigger quality.

Suggestions

Add an explicit 'Use when...' clause stating when to invoke the skill (e.g. 'Use when writing AE/ICSR narratives for FDA, EMA, or CIOMS I submission, or when the user mentions AE narrative, ICSR, safety report, or MedWatch narrative').

Include conversational trigger variations users would naturally say ('AE narrative', 'safety report', 'MedWatch narrative') alongside the formal regulatory terms.

Dimension	Reasoning	Score
Specificity	Lists multiple specific concrete actions — "Generates CIOMS I-compliant ICSR narratives", "temporal analysis", "MedDRA coding", "causality assessment using WHO-UMC or Naranjo criteria", "multi-format output" — rather than vague language, matching the 'Lists multiple specific concrete actions' anchor.	3 / 3
Completeness	Clearly answers 'what' (generates narratives, temporal analysis, MedDRA coding, causality, multi-format output) but has no explicit 'when'/'Use when...' trigger clause; per the guidelines, a missing 'Use when...' clause caps completeness at 2.	2 / 3
Trigger Term Quality	Contains relevant formal domain keywords ("ICSR", "CIOMS", "MedDRA", "causality assessment", "FDA", "EMA") but misses the conversational variations a user might actually say (e.g. "AE narrative", "safety report", "MedWatch narrative") that appear only in the body, fitting 'Some relevant keywords but missing common variations'.	2 / 3
Distinctiveness Conflict Risk	The CIOMS I-compliant ICSR / FDA-EMA regulatory submission niche is highly specific with distinct domain triggers, making it unlikely to fire for the wrong skill — matching 'Clear niche with distinct triggers; unlikely to conflict'.	3 / 3
	Total	10 / 12 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
Commit: d924410

Reviewed: 3 days 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.