The Agent Skills Directory

[PROMPT_INJECTION] (MEDIUM): The skill is vulnerable to indirect prompt injection from the codebase being analyzed. \n
Ingestion points: The skill uses git diff and git diff --staged to read external file content into the agent context. \n
Boundary markers: No explicit delimiters or instructions to ignore embedded commands within the diffs are provided. \n
Capability inventory: The skill has the authority to run git add and git commit, which modifies the filesystem and repository history. \n
Sanitization: No sanitization of the diff content is performed before the agent uses it to generate text for the commit message. \n- [COMMAND_EXECUTION] (LOW): The skill relies on the Bash tool to execute git commands. \n
Evidence: The workflow uses shell execution for all operations. While restricted to git commands, a malicious diff could attempt to trick the agent into generating a commit message that includes shell metacharacters to escape the git commit -m command.

git-commit