The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses git status and git log to ground its summary in the actual repository state. These are standard, safe commands for a development-focused agent skill.
[INDIRECT_PROMPT_INJECTION]: The skill has a surface for indirect injection as it processes conversation history and git logs. • Ingestion points: The skill logic reads the session history and git log command outputs. • Boundary markers: Uses a specific markdown template for recap construction. • Capability inventory: Limited to read-only information gathering and display; no file-write or network-out capabilities are used in conjunction with this ingestion. • Sanitization: No explicit sanitization of log content is performed, but the impact is contained within the generated summary.

recap