The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute standard local Git commands, including git status, git branch, git rev-parse, and git worktree. These are used for their intended purpose of repository inspection. The only write operation permitted is a git fetch without pruning, which updates local remote-tracking references but does not modify the working tree or branch state.
[INDIRECT_PROMPT_INJECTION]: The skill has an indirect prompt injection surface as it processes data from external Git providers (Pull Requests, Issues) and local repository metadata (branch names, commit messages). 1. Ingestion points: Responses from Git provider APIs and output from local Git commands in the current workspace. 2. Boundary markers: None specified; the skill does not explicitly instruct the agent to treat this data as untrusted literals. 3. Capability inventory: The skill is restricted to file system inspection via Git and read-only network calls to Git providers. 4. Sanitization: No explicit sanitization or validation steps are defined for the ingested repository data.

kata-sitrep