The Agent Skills Directory

[SAFE]: No malicious patterns such as prompt injection, data exfiltration, or obfuscation were detected in the skill's instructions or metadata.
[COMMAND_EXECUTION]: The skill instructs the agent to run git diff --cached and git diff --cached --name-status. These are standard, read-only git commands required for the skill to perform its primary function and do not pose a security risk in this context.
[PROMPT_INJECTION]: The skill processes untrusted content (staged code diffs) which represents an indirect prompt injection surface. However, the risk is negligible as the agent's output is strictly constrained to a specific commit message format and the skill lacks access to sensitive capabilities like network operations or file writes.
Ingestion points: Staged git changes are read via git diff --cached.
Boundary markers: The skill does not use specific delimiters to encapsulate the diff data.
Capability inventory: The skill is limited to reading diffs and generating text; it does not utilize tools for network access, arbitrary command execution, or file modifications.
Sanitization: No sanitization is performed on the diff content before analysis.

git-commit-generator