Defuddle (kepano)

Extract clean markdown content from web pages using Defuddle CLI, removing clutter and navigation to save tokens. Use instead of WebFetch when the user provi...

Security Scan

The skill's instructions and requirements are coherent with its stated purpose (extracting markdown from web pages), but the metadata omits declaring the required defuddle CLI and the README suggests installing an npm package — you should verify that package before installing it globally.

Purpose & Capability

The SKILL.md clearly describes using the Defuddle CLI to extract markdown from web pages, which matches the skill name/description. However, the registry metadata lists no required binaries while the instructions require the 'defuddle' CLI (and suggest 'npm install -g defuddle'), so the metadata is missing a declared dependency.

Instruction Scope

Instructions are narrowly scoped to running the defuddle CLI (parse <url> --md and related flags) and saving output; they do not ask the agent to read unrelated files, access extra environment variables, or transmit data to third-party endpoints beyond fetching the provided URL.

Install Mechanism

There is no formal install spec in the skill (instruction-only). SKILL.md recommends installing via 'npm install -g defuddle' — this is a reasonable, common mechanism but relies on an external npm package. Installing packages from public registries carries moderate risk unless the package and publisher are verified.

Credentials

The skill requests no environment variables, credentials, or config paths. That is proportionate for a wrapper around a CLI tool that operates on provided URLs.

Persistence & Privilege

Skill does not request always:true and is user-invocable with standard autonomous invocation allowed. It does not request persistent system-wide privileges or modify other skills' configuration.

Assessment

This skill is coherent: it simply instructs the agent how to use the Defuddle CLI to turn web pages into markdown. Before installing or using it, verify the 'defuddle' npm package and its publisher (npm packages can contain arbitrary code). Consider one of: (1) pre-installing a vetted version of defuddle in the agent environment and add that binary to the agent's allowed list, (2) confirm the npm package's trustworthiness and review its source, or (3) ask the skill author to declare the required binary in the skill metadata so dependency checks can run automatically. If you do not want the agent to run arbitrary global installs or execute external CLIs, do not grant it permission to run the suggested install command.

Defuddle

Use Defuddle CLI to extract clean readable content from web pages. Prefer over WebFetch for standard web pages — it removes navigation, ads, and clutter, reducing token usage.

If not installed: npm install -g defuddle

Usage

Always use --md for markdown output:

defuddle parse <url> --md

Save to file:

defuddle parse <url> --md -o content.md

Extract specific metadata:

defuddle parse <url> -p title
defuddle parse <url> -p description
defuddle parse <url> -p domain

Output formats

Flag	Format
--md	Markdown (default choice)
--json	JSON with both HTML and markdown
(none)	HTML
-p <name>	Specific metadata property