This is a strong, well-crafted description that thoroughly covers capabilities, trigger conditions, and edge cases. It uses concrete action verbs, includes natural user language, and explicitly defines when the skill should activate. The only minor note is the use of second-person 'Use this skill' framing, but the bulk of the description uses imperative/third-person action language, and the second-person usage is in the context of instructing Claude rather than addressing the user.

Dimension	Reasoning	Score
Specificity	Lists multiple specific concrete actions: creating slide decks/pitch decks/presentations, reading/parsing/extracting text, editing/modifying/updating, combining/splitting slide files, working with templates/layouts/speaker notes/comments.	3 / 3
Completeness	Clearly answers both 'what' (creating, reading, editing, combining presentations) and 'when' with explicit triggers ('Trigger whenever the user mentions deck, slides, presentation, or references a .pptx filename'). The entire description is essentially a detailed 'Use when...' clause.	3 / 3
Trigger Term Quality	Excellent coverage of natural terms users would say: 'deck,' 'slides,' 'presentation,' '.pptx,' 'pitch decks,' 'speaker notes,' 'templates,' and covers scenarios like extracting content for use in emails or summaries.	3 / 3
Distinctiveness Conflict Risk	Clearly scoped to .pptx files with distinct triggers like 'deck,' 'slides,' 'presentation,' and '.pptx filename.' The description even clarifies edge cases (e.g., extracting content for use elsewhere still triggers this skill), reducing ambiguity and conflict risk.	3 / 3
	Total	12 / 12 Passed

Implementation

85%

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

This is a well-structured, highly actionable skill with excellent workflow clarity and progressive disclosure. The QA section is particularly strong with its explicit verification loops, subagent delegation, and grep-based validation. The main weakness is the lengthy Design Ideas section, which contains substantial general design guidance (color palettes, typography rules, layout patterns) that inflates token count — though it does provide project-specific value like the 'NEVER use accent lines under titles' anti-pattern.

Suggestions

Consider moving the Design Ideas section (color palettes, typography, layout options) to a separate design.md file and referencing it from the main skill, keeping only the most critical anti-patterns inline.

Dimension	Reasoning	Score
Conciseness	The quick reference table and reading/editing sections are lean and efficient. However, the Design Ideas section is very lengthy (~100 lines of color palettes, typography tables, and layout advice) — much of which is general design knowledge Claude likely already possesses. The color palette table and font pairing table add significant token weight.	2 / 3
Actionability	The skill provides fully executable commands (markitdown, thumbnail.py, unpack.py, soffice.py, pdftoppm), concrete QA grep patterns, specific subagent prompts with exact inspection criteria, and clear verification loops. The guidance is copy-paste ready throughout.	3 / 3
Workflow Clarity	The QA section has an explicit verification loop with feedback (generate → inspect → list issues → fix → re-verify → repeat). The editing workflow references a detailed guide. The 'Do not declare success until you've completed at least one fix-and-verify cycle' is a strong validation checkpoint. The content QA includes grep-based validation for leftover placeholders.	3 / 3
Progressive Disclosure	The quick reference table immediately routes to the right guide (editing.md, pptxgenjs.md) based on task type. The SKILL.md serves as a clear overview with one-level-deep references. Content is well-organized into distinct sections (Reading, Editing, Creating, Design, QA, Converting, Dependencies).	3 / 3
	Total	11 / 12 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 — 11 / 11 Passed

Validation for skill structure

No warnings or errors.

Reviewed

12 days ago

Table of Contents

Discovery Implementation Validation