The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because its core workflow depends on reading and following instructions from external, untrusted files found within a repository.
Ingestion points: The agent is explicitly instructed in SKILL.md to read CONTRIBUTING, CONTRIBUTING.md, or similar files to extract commit message standards and requirements. It also performs git diff on modified files.
Boundary markers: There are no boundary markers or instructions provided to the agent to treat the content of these files as data rather than instructions. The agent is likely to obey instructions embedded within these files.
Capability inventory: The skill utilizes shell command execution via the git CLI (e.g., git status, git diff, git add, git commit).
Sanitization: The skill lacks any sanitization or validation logic for the content extracted from repository files.
[COMMAND_EXECUTION]: The skill relies on executing shell commands to interact with the git version control system. While these are necessary for the skill's functionality, they represent a capability that could be abused if the agent is successfully injected with malicious instructions from the repository it is analyzing.

git-commit-workflow