pptx-skill

Triggers whenever the user asks to do anything with PowerPoint slides (.pptx): read, summarize, extract, edit, create, or convert. Key insight: PPTX files are visually structured — charts, diagrams, and layouts cannot be reliably captured by text extraction alone. The correct approach for understanding slide content is to render the target slide(s) as PNG images (via PptxInspect render) and pass them to a vision model (CloudImageUnderstand), not to rely solely on text/outline extraction. Use text tools (outline / text / shapes) only as supplementary context. Typical phrasings: "summarize this PPT", "what is this deck about", "what's on slide 3", "describe page 5", "make me a Q4 report deck", "change the title of slide 3". Load this skill whenever the user mentions a .pptx file, deck, slides, or presentation.

Quality

88%

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

Quality

Content

76%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 body is highly actionable with multiple full executable demos, concrete tool calls, decision tables, and a specific error-recovery table, and it is mostly token-efficient and well-structured. Its main weakness is workflow clarity: the destructive XML-editing workflow lacks a validation checkpoint, and the two overlapping generation demos plus inlined code could be trimmed or moved to reference files.

Suggestions

Add a validation checkpoint after the adm-zip XML patch in §2.3 — e.g., re-open the deck with PptxInspect(summary) or render the edited slide to confirm it parses and displays before declaring success.

Consolidate or shorten the second generation demo (§2.2) since §2.1 already demonstrates the pptxgenjs pattern, or move full demo code into a references/ file to improve token efficiency and progressive disclosure.

Add an explicit verify step to the batch-rendering workflow beyond the >20-slide user confirmation, such as spot-checking one rendered PNG before processing the whole deck.

Dimension	Reasoning	Score
Conciseness	Mostly efficient and assumes Claude's competence (no generic explanations of what PPTX is or how libraries work), but two overlapping full generation demos (§2.1 and §2.2) and inlined large code blocks could be trimmed, so not every token earns its place.	4 / 5
Actionability	Fully executable, copy-paste-ready code covering generation, multi-slide generation, XML editing, plus concrete PptxInspect/DocConvert calls, decision tables, and a specific error-recovery table covering the common cases.	5 / 5
Workflow Clarity	Sequences are present (read → write → convert with branching) and there are some checkpoints (anti-hallucination rule, >20-slide user confirmation), but the destructive adm-zip XML patch in §2.3 has no validation/verification step, capping workflow clarity per the rubric.	3 / 5
Progressive Disclosure	Good section structure with clear navigation (Tool List, Read, Write, Visual, Convert, Error Recovery) and no nested references; the minor gap is that large demo code is inlined rather than split into reference files, though no bundle files exist to point to.	4 / 5
	Total	16 / 20 Passed

Description

100%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 lists concrete capabilities, provides comprehensive natural trigger terms with the .pptx extension and synonyms, and explicitly states both what and when in third-person voice. It is somewhat verbose — the 'Key insight' and rendering-approach guidance read more like body content than a trigger description — but this does not undermine the four scored dimensions.

Dimension	Reasoning	Score
Specificity	Lists six concrete capabilities — 'read, summarize, extract, edit, create, or convert' — giving comprehensive coverage of actions rather than vague language, matching the score-5 anchor.	5 / 5
Completeness	Explicitly answers both 'what' (the six actions) and 'when' ('Load this skill whenever the user mentions a .pptx file, deck, slides, or presentation') with concrete trigger phrases, so no missing-trigger cap applies.	5 / 5
Trigger Term Quality	Comprehensive natural-term coverage including the '.pptx' extension and synonyms ('PPT', 'deck', 'slides', 'presentation') plus concrete typical phrasings users would actually say.	5 / 5
Distinctiveness Conflict Risk	PowerPoint/.pptx is a clear niche with distinct triggers and minimal overlap risk with other skills; the file-extension anchor makes misfiring unlikely.	5 / 5
	Total	20 / 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: itsablabla/openloaf-web
Path: apps/server/src/ai/builtin-skills/pptx/en/SKILL.md
Commit: a1ab5be

Reviewed: 4 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.