The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill metadata and instructions facilitate the installation of a Go binary from a third-party GitHub repository (github.com/Hyaxia/blogwatcher). This source is not associated with a known trusted organization or well-known service.
[COMMAND_EXECUTION]: The skill's primary functionality involves executing the blogwatcher command-line interface. The agent is directed to run multiple commands including blogwatcher add, blogwatcher scan, and blogwatcher articles, which perform network operations to fetch external data and manage local state.
[PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection because it retrieves and displays content from external websites.
Ingestion points: Untrusted data enters the agent's context through RSS/Atom feeds processed by the blogwatcher scan and blogwatcher articles commands.
Boundary markers: The skill lacks explicit delimiters or instructions to prevent the agent from obeying commands that might be embedded within the feed content.
Capability inventory: The skill utilizes shell command execution via the blogwatcher CLI and file system interactions for managing tracked blogs.
Sanitization: There is no evidence of sanitization, filtering, or validation performed on the external content before it is presented to the agent.

blogwatcher