The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands, specifically Git tools like git log and git diff, to analyze repository history and identify changes for patch creation.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection via micropatch.md files. These files are defined as the authoritative source of truth for the agent's actions during the implementation and verification phases. A maliciously crafted patch file could contain instructions designed to trick the agent into performing unauthorized code modifications or executing harmful commands.\n
Ingestion points: Processes content from micropatch.md and hints.diff files provided as input to the skill.\n
Boundary markers: There are no explicit delimiters or instructions to treat the external file content as data rather than instructions.\n
Capability inventory: The agent has the capability to modify the filesystem, execute Git commands, and run verification steps defined in the 'Definition of Done'.\n
Sanitization: No validation or sanitization is performed on the instructions, code snippets, or verification commands contained within the patch files.

micropatch