ectd-xml-compiler

Automatically convert uploaded drug application documents (Word/PDF) into XML skeleton structure compliant with eCTD 4.0/3.2.2 specifications.

Quality

40%

Does it follow best practices?

Impact

—

No eval scenarios have been run

Securityby

Passed

No known issues

Fix and improve this skill with Tessl

tessl review fix ./scientific-skills/Academic Writing/ectd-xml-compiler/SKILL.md

Quality

Content

37%

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

The skill documents a genuine tool with executable entry points and a reference file, but the body is bloated with boilerplate and explanatory padding, has duplicate/broken sections, and lacks the enforced validation feedback loop its XML-generation workflow requires.

Suggestions

Cut the explanatory Overview/eCTD-structure directory tree and generic risk/security/evaluation/response-template boilerplate; keep only what Claude would not already know and move detailed module-mapping rules into a reference file.

Make validation an enforced checkpoint in the workflow (generate -> validate -> if errors fix and re-validate -> only proceed when valid) instead of an optional flag, to satisfy the feedback-loop requirement for XML/batch operations.

Resolve the duplicate 'References' sections and the placeholder 'See ## Usage above' / 'See ## Workflow above' stubs, and reconcile the contradictory 'Prerequisites: No additional Python packages required' with the stated Dependencies.

Dimension	Reasoning	Score
Conciseness	The body runs ~310 lines and is padded with content Claude already knows (an 'Overview' explaining what eCTD/ICH/FDA/EMA are, an eCTD directory tree, risk-assessment/security/evaluation tables, a response template, and boilerplate duplicated from a generic template), plus duplicate 'References' sections and contradictory 'Prerequisites: No additional Python packages required' against a real Dependencies block.	1 / 3
Actionability	It provides concrete executable commands (py_compile, --help, pip install, a usage line with a real options table) and a packaged main.py, but the core conversion logic is described only as keyword-matching rules and document-parsing steps rather than shown as copy-paste ready code, and several example blocks contain placeholder/blank lines.	2 / 3
Workflow Clarity	A numbered workflow and validation step exist, but for an XML-generation/batch operation the rubric requires an explicit validate->fix->retry feedback loop; validation is only mentioned as an optional '--validate' flag with no checkpoint enforcing 'only proceed when valid', so workflow clarity is capped at 2.	2 / 3
Progressive Disclosure	A real references/audit-reference.md is linked one level deep and scripts/main.py is the implementation surface, but the body keeps large blocks inline that belong in references (eCTD structure tree, module auto-recognition rules, risk/security tables), duplicates the References section, and includes broken cross-references ('See ## Usage above', 'See ## Workflow above' as standalone sections).	2 / 3
	Total	7 / 12 Passed

Description

42%

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 conveys what the skill does and occupies a distinct niche, but it is a single sentence with no explicit trigger guidance, so it fails to tell Claude when to use it.

Suggestions

Add an explicit 'Use when...' clause naming natural user phrasings, e.g. 'Use when converting drug application Word/PDF documents into eCTD-compliant XML, or when preparing eCTD/regulatory submission skeletons.'

List multiple concrete actions (e.g. parse document heading hierarchy, map sections to eCTD modules m1-m5, generate index.xml and MD5 checksums, validate against the DTD) to raise specificity from a single action to several.

Include common natural trigger terms a user would actually say ('eCTD submission', 'regulatory filing', 'drug registration XML').

Dimension	Reasoning	Score
Specificity	Names the domain ('drug application documents (Word/PDF)') and the concrete action ('convert... into XML skeleton structure'), but lists only one action rather than multiple specific actions like the score-3 anchor requires.	2 / 3
Completeness	Answers 'what' clearly but is missing any 'Use when...' or equivalent explicit 'when' guidance, which the rubric states caps completeness at 2; it only describes capability, not when to invoke it.	1 / 3
Trigger Term Quality	Includes relevant terms ('drug application documents', 'Word', 'PDF', 'eCTD', 'XML'), but lacks common natural variations a user might say ('eCTD submission', 'regulatory filing', 'drug registration') and has no explicit trigger clause.	2 / 3
Distinctiveness Conflict Risk	The eCTD/regulatory-submission niche is highly specific and unlikely to conflict with other skills; the combination of 'eCTD', 'drug application', and 'XML skeleton' is distinctive.	3 / 3
	Total	8 / 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: 7cc5680

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.