The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses shell commands like git log and git config to extract repository information and user identity. This is part of the core functionality but involves direct command execution.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted commit messages from the local Git history, which serves as an injection surface.
Ingestion points: Data is collected via git log in references/execution-flow.md.
Boundary markers: The instructions do not define delimiters to separate commit data from agent instructions during the analysis phase.
Capability inventory: The skill has the capability to execute git commands and perform file/directory analysis.
Sanitization: There is no evidence of sanitization or filtering of commit message content before it is processed by the AI.

weekly-report