The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes data from external sources including issue specifications, pull request metadata, and codebase content, which creates a surface for indirect instructions. \n- Ingestion points: SKILL.md (instructions to read issues and specifications), references/epic-simplification.md (instructions to read parent specifications and integrated PR inventory). \n- Boundary markers: The skill relies on natural language instructions to follow project specifications and AGENTS.md but lacks formal delimiters to separate untrusted data from the agent's core instructions. \n- Capability inventory: The skill has access to shell utilities (git, rg) and the ability to modify files within the repository scope. \n- Sanitization: There are no explicit requirements for sanitizing or escaping the content of processed issues or code. \n- [COMMAND_EXECUTION]: The skill instructs the agent to use local command-line tools for repository analysis. \n- Evidence: SKILL.md specifies the use of git status --short, git diff, and git ls-files. references/epic-simplification.md specifies the use of rg (ripgrep) for code searches.

reduce-complexity