The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data by reading git diffs and the contents of untracked files to generate summaries. This creates a surface where malicious instructions hidden in source code or documentation being committed could influence the agent's behavior.
Ingestion points: The skill reads output from git diff, git status, and file contents using cat and sed (SKILL.md).
Boundary markers: Absent. There are no instructions to the agent to treat the ingested content as strictly data or to ignore instructions embedded within the files.
Capability inventory: The skill uses shell-based tools including git, cat, and sed (SKILL.md).
Sanitization: Absent. The skill does not implement any filtering or escaping of the content read from the file system before it is processed by the model.
[COMMAND_EXECUTION]: The skill instructions direct the agent to execute multiple shell commands, including git status, git diff, cat, and sed. While these are functional requirements for generating commit messages, they grant the agent the ability to read and process arbitrary file content within the repository.

git-committer