The Agent Skills Directory

Indirect Prompt Injection (LOW): The skill requires the agent to process output from git diff commands, which is untrusted external data. Maliciously crafted code comments or changes could include instructions aimed at influencing the agent's behavior or output. 1. Ingestion points: Data from git diff --staged commands. 2. Boundary markers: Absent; no specific delimiters are used to separate diff content from instructions. 3. Capability inventory: The skill is restricted to generating text and does not perform file modifications, network requests, or arbitrary code execution. 4. Sanitization: No sanitization or filtering of the diff content is mentioned or implemented.

commit-message-generator