Content
17%Weight 40%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The body reads as a product landing page rather than an operational skill: it promotes a Python pipeline it does not ship, provides zero executable instructions, and pads token budget with FAQ and marketing copy. Claude loading this skill would learn what the tool supposedly does but have no idea how to do any of it.
Suggestions
Ship the pipeline scripts (or inline their commands) and replace the 'What it does' descriptions with concrete invocations, e.g. 'python scripts/extract.py ./dist' followed by generation and scoring commands.
Delete the marketing prose ('Why use this skill?', 'Turn Your Website into an Agent-Ready Storefront') and the FAQ section; retain only requirements, deliverable file formats, and the worked example.
Add an explicit step sequence with validation checkpoints (e.g. validate generated JSON-LD against schema.org before writing the agent-ready/ directory) and move schema templates and readiness-scoring criteria into referenced files under references/.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Marketing framing ("As AI agents like Claude, ChatGPT, and shopping bots begin to navigate the web autonomously, traditional HTML isn't enough") and a six-question FAQ re-explaining how agents read websites pad the body with concepts Claude already knows. Not 1 because a few genuinely useful facts survive (Python 3.11+ requirement, deliverables, known limitations). | 2 / 5 |
Actionability | The body only describes the pipeline ("The skill executes a multi-stage pipeline using Python-based extraction and generation scripts") with no commands, code, script paths, or API usage, and it explicitly states the scripts "are not in this folder". There is nothing executable for Claude to act on, so it cannot reach 2's 'high-level hints with specific steps missing'. | 1 / 5 |
Workflow Clarity | The stages (Extraction, Generation, Formatting, Readiness Scoring) appear only as capability descriptions, not as an executable sequence, and no validation or verification steps exist for what is a batch web-scraping operation. Not 3 because there is no actionable step sequence at all, only a rough conceptual order with validation absent. | 2 / 5 |
Progressive Disclosure | No bundle files exist and the body references none, while FAQ and promotional content that belongs outside SKILL.md is inlined. Not 3 because the section structure present is navigational marketing rather than skill organization, and the detailed material the skill needs (pipeline scripts, schema templates) exists nowhere to be disclosed. | 2 / 5 |
Total | 7 / 20 Passed |