The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes the entire CLI and git commands to manage review trails, check branches, and apply modifications to the codebase.\n- [EXTERNAL_DOWNLOADS]: Retrieves finding data, descriptions, and code patches from entire.io. This is the vendor's official domain and is necessary for the skill to function as intended.\n- [REMOTE_CODE_EXECUTION]: The skill fetches and applies unified-diff patches from a remote source. While this is the core functionality for addressing findings, it allows remote content to modify the local codebase.\n- [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface because it processes natural language finding bodies from external sources.\n
Ingestion points: Finding descriptions and metadata fetched via entire trail finding list in SKILL.md (Step 3).\n
Boundary markers: Absent; the agent processes finding bodies directly without delimiters or instructions to ignore embedded commands.\n
Capability inventory: File modification and execution of the entire CLI tool as defined across SKILL.md.\n
Sanitization: No validation or sanitization is performed on the content of the findings before the agent processes them.

address-findings