The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the installation of an external package from a source that is not listed as a trusted vendor. Specifically, it recommends running npm install -g defuddle if the tool is not present. This introduces a supply chain risk as the integrity of this third-party package is not verified within the skill.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to fetch and process untrusted content from arbitrary external URLs, which creates an attack surface where malicious instructions embedded in a webpage could manipulate the agent's behavior.
Ingestion points: External web content is ingested via the defuddle parse <url> command from user-provided URLs.
Boundary markers: The skill does not provide instructions to the agent on how to differentiate between the fetched content and its own instructions (e.g., using XML tags or delimiters).
Capability inventory: The skill allows the agent to execute the defuddle CLI tool and write output to the local file system (e.g., -o content.md).
Sanitization: There is no evidence of sanitization, filtering, or validation of the fetched content before it is returned to the agent's context.

defuddle