The Agent Skills Directory

[COMMAND_EXECUTION]: The scripts scripts/generate_commit_message.py and scripts/suggest_commit_type.py use the subprocess module to execute git diff commands. This is used to programmatically retrieve code changes for analysis and is inherent to the skill's primary function.
[PROMPT_INJECTION]: The skill processes external, untrusted data (git diffs) that could contain malicious instructions designed to influence the AI's output (indirect prompt injection).
Ingestion points: Untrusted diff content is ingested through scripts/generate_commit_message.py and scripts/suggest_commit_type.py via shell output or file reads.
Boundary markers: The skill does not implement explicit boundary markers or instructions to isolate the diff content from the agent's instructions.
Capability inventory: The skill is configured with allowed-tools: Bash(cmd:*), granting significant execution privileges to the agent.
Sanitization: No content sanitization is performed on the diff data before it is presented to the model for analysis.

generating-conventional-commits