CtrlK
BlogDocsLog inGet started
Tessl Logo

defuddle

Extract clean markdown content from web pages using Defuddle CLI, removing clutter and navigation to save tokens. Use instead of WebFetch when the user provides a URL to read or analyze, for online documentation, articles, blog posts, or any standard web page.

69

Quality

84%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

85%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.

A highly actionable, well-structured skill with executable commands, a clear fallback workflow, and a properly signaled one-level-deep reference file. The main weakness is redundancy: the fallback rule and mental-model framing are repeated across three sections and could be consolidated for token efficiency.

Suggestions

State the fallback rule ("if output is empty or navigation-only, fall back to WebFetch") once — in Quick Start — and remove the near-verbatim repeats in the Mindset section and the Mental model paragraph.

Delete or drastically shorten the 'Mental model' paragraph, which restates the decision table immediately above it; the table already carries the same information.

Merge the 'Mindset' and 'When to Use' sections into one, keeping the WHEN/WHEN-NOT bullets and the table while dropping the repeated prose framing.

DimensionReasoningScore

Conciseness

The body is mostly efficient (no explaining of concepts Claude already knows; commands given directly), but the guidance "When in doubt, try defuddle first; if the output is empty or only navigation text, fall back to WebFetch" is repeated nearly verbatim three times (Mindset, Quick Start step 3, and the Mental model paragraph), and the Mental model paragraph restates the preceding decision table. This matches anchor 3 (mostly efficient but could be tightened) rather than anchor 4, where only minor trims would be needed.

3 / 5

Actionability

Commands are copy-paste ready and cover the common cases: "defuddle parse <url> --md", "defuddle parse <url> --md -o content.md", "defuddle parse <url> -p title", plus install ("npm install -g defuddle") and paired BAD/GOOD examples for every mistake. This fully matches anchor 5; anchor 4 would imply gaps in coverage of common cases.

5 / 5

Workflow Clarity

Quick Start gives a clear sequence (confirm page type → run parse → check output → save) with an explicit validation checkpoint and error-recovery loop: "If output is empty or only navigation text, fall back to WebFetch". The operation is non-destructive so no further validation is required, matching anchor 5's feedback-loop pattern.

5 / 5

Progressive Disclosure

The single bundle file (references/ANTI-PATTERNS.md) exists, is referenced exactly once with a clear signal ("See [ANTI-PATTERNS.md](references/ANTI-PATTERNS.md) for full explanations of each mistake"), and is one level deep with no nested references. The split is appropriate: BAD/GOOD commands stay inline for immediate visibility while the WHY explanations live in the reference, matching anchor 5.

5 / 5

Total

18

/

20

Passed

Description

83%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.

A strong description that clearly states capability and explicit trigger conditions, and even positions itself against the nearest alternative (WebFetch). Minor improvements possible: mention the metadata-extraction and save-to-file capabilities, and add a few more natural synonyms like "link" or "webpage".

DimensionReasoningScore

Specificity

"Extract clean markdown content from web pages using Defuddle CLI, removing clutter and navigation to save tokens" lists several concrete actions (extract markdown, strip clutter/navigation, reduce token usage), but coverage has minor gaps — metadata extraction (-p) and file output (-o) capabilities are not mentioned. It sits above anchor 3 (only 1-2 actions) and just below anchor 5 (comprehensive coverage).

4 / 5

Completeness

Explicitly answers what ("Extract clean markdown content from web pages using Defuddle CLI, removing clutter and navigation") and when ("Use instead of WebFetch when the user provides a URL to read or analyze, for online documentation, articles, blog posts, or any standard web page") with concrete trigger phrases. This matches the anchor-5 example pattern almost exactly; anchor 4 would require the 'when' clause to be less explicit.

5 / 5

Trigger Term Quality

Triggers include "when the user provides a URL to read or analyze", "online documentation, articles, blog posts, or any standard web page" — natural phrases users would say. A few common variants are missing ("link", "webpage", "wiki", "website"), so it does not reach the comprehensive-synonym coverage of anchor 5.

4 / 5

Distinctiveness Conflict Risk

"Use instead of WebFetch" explicitly differentiates it from the closest competing tool, giving a clear niche. However, "any standard web page" is broad enough to overlap with general web-fetching situations, keeping it at anchor 4 rather than the minimal-conflict anchor 5.

4 / 5

Total

17

/

20

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.

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
pantheon-org/tekhne
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.