The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes untrusted data from the output of git commands, creating an indirect prompt injection surface where malicious comments or code in the diff could attempt to influence the agent.
Ingestion points: The skill reads data from the local environment via git diff --staged (SKILL.md).
Boundary markers: No specific delimiters or instructions are provided to the agent to treat the diff content as data rather than instructions.
Capability inventory: The skill involves reading repository state and formatting text; it explicitly requires user confirmation before committing changes.
Sanitization: There is no evidence of filtering or sanitizing the diff content before it is processed by the agent.
[COMMAND_EXECUTION]: The skill executes git diff --staged to inspect the local repository. This is a legitimate operation for a development tool and does not involve high-privilege escalation.

gerar-commit