web-to-markdown
Use ONLY when the user explicitly says: 'use the skill web-to-markdown ...' (or 'use a skill web-to-markdown ...'). Converts webpage URLs to clean Markdown by calling the local web2md CLI (Puppeteer + Readability), suitable for JS-rendered pages.
82
81%
Does it follow best practices?
Impact
74%
1.68xAverage score across 3 eval scenarios
Advisory
Suggest reviewing before use
Quality
Discovery
85%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This description is technically complete and highly distinctive, with clear what/when clauses. However, the explicit invocation requirement severely limits discoverability - users must know the exact skill name to trigger it, which defeats the purpose of intelligent skill selection. The description would benefit from natural trigger terms that allow Claude to proactively suggest this skill when users mention web scraping or URL conversion.
Suggestions
Add natural trigger terms users would say, such as 'scrape webpage', 'convert URL to markdown', 'fetch website content', 'extract page text'
Consider relaxing the 'ONLY when explicitly says' constraint to allow Claude to select this skill when users naturally describe web-to-markdown conversion needs
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists concrete actions: 'Converts webpage URLs to clean Markdown by calling the local web2md CLI (Puppeteer + Readability), suitable for JS-rendered pages.' This describes specific technology and use case. | 3 / 3 |
Completeness | Clearly answers both what ('Converts webpage URLs to clean Markdown by calling the local web2md CLI') and when ('Use ONLY when the user explicitly says: use the skill web-to-markdown'). The when clause is explicit, even if restrictive. | 3 / 3 |
Trigger Term Quality | Includes some relevant terms like 'webpage URLs', 'Markdown', 'web2md', but the explicit invocation requirement ('use the skill web-to-markdown') limits natural trigger coverage. Missing common variations like 'scrape', 'fetch page', 'convert website'. | 2 / 3 |
Distinctiveness Conflict Risk | Very distinct due to the explicit invocation requirement and specific technology stack (web2md CLI, Puppeteer + Readability). Unlikely to conflict with other skills given the narrow trigger condition. | 3 / 3 |
Total | 11 / 12 Passed |
Implementation
77%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, actionable skill with clear workflows and validation steps. The main weakness is moderate verbosity in explaining optional flags and browser compatibility details that could be trimmed or moved to a reference file. The hard trigger gate section is appropriately prominent and the executable examples are excellent.
Suggestions
Move the extensive optional rendering controls list to a separate REFERENCE.md file, keeping only the most common flags inline
Trim the 'What this skill does' section - Claude doesn't need explanation of Puppeteer/Readability/Turndown concepts
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is reasonably efficient but includes some unnecessary explanation (e.g., 'Works best with Chromium-family browsers' and the full list of optional flags could be trimmed). The 'What this skill does' section explains things Claude could infer from the tool name and usage. | 2 / 3 |
Actionability | Provides fully executable, copy-paste ready commands for all scenarios (single URL, batch, interactive mode, stdout). Installation instructions are concrete with exact paths and commands. | 3 / 3 |
Workflow Clarity | Clear 6-step workflow with explicit validation checkpoints (step 3 checks installation, step 5 validates output with specific commands like 'ls -la' and 'wc -c'). Includes error recovery path for missing tool installation. | 3 / 3 |
Progressive Disclosure | Content is well-organized with clear sections, but the optional rendering controls list is lengthy and could be moved to a reference file. For a skill of this size (~80 lines), inline content is acceptable but borders on verbose. | 2 / 3 |
Total | 10 / 12 Passed |
Validation
100%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.
- Repository
- softaworks/agent-toolkit
- Commit
3027f20
- Reviewed
Table of Contents
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.