The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes several shell commands to automate skill management, including npx to run the skills CLI, gh to retrieve the current GitHub user, and git to access global configuration settings. These are necessary for its defined workflow.
[EXTERNAL_DOWNLOADS]: The skill downloads and installs agent skills from GitHub repositories. It focuses on the user's own repositories and any specific sources provided in the $ARGUMENTS field. This is the primary function of the skill.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from external sources and incorporates it into shell commands without sanitization, creating a potential vulnerability surface.
Ingestion points: The skill ingests data from the GitHub API, Git configuration, user-supplied arguments, and the console output of the skills update command in SKILL.md.
Boundary markers: There are no boundary markers or instructions to the agent to validate or ignore potentially malicious content within the external data being processed.
Capability inventory: The skill possesses significant capabilities, including executing packages via npx and interacting with the file system and network via the skills CLI tool.
Sanitization: No sanitization or validation logic is present to filter out malicious payloads that might be embedded in repository names or CLI tool outputs before they are used as arguments in subsequent commands.

sync-skills