The Agent Skills Directory

[COMMAND_EXECUTION]: The skill invokes the git CLI tool to perform repository management tasks including fetch, status, log, and pull.
[EXTERNAL_DOWNLOADS]: The skill is designed to retrieve updates from a remote Git repository (origin). This introduces a dependency on the integrity of the remote source, which is not restricted to a trusted vendor.
[REMOTE_CODE_EXECUTION]: By pulling and applying updates to the local workspace, the skill can modify its own instructions and executable scripts. If the remote repository is compromised, this mechanism could be used to deploy malicious updates to the agent's environment.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface through its update mechanism.
Ingestion points: The git pull operation in SKILL.md imports external content into the agent's workspace.
Boundary markers: There are no explicit delimiters or instructions to the agent to disregard instructions embedded within commit messages or updated files.
Capability inventory: The skill possesses file-write capabilities and command execution through the git toolset.
Sanitization: Content pulled from the remote is not validated or sanitized before being processed.
[PERSISTENCE]: The skill utilizes a schedule tool to establish recurring update checks (e.g., 0 9 * * *), ensuring the update logic executes automatically across sessions.

auto-update