The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external codebases and pull requests which could contain malicious instructions designed to influence the agent's analysis or actions.
Ingestion points: The skill ingests data from git diff output, rg (ripgrep) results, and project source files (TypeScript, JSON, etc.) across the monorepo.
Boundary markers: The skill lacks explicit boundary markers or instructions to treat analyzed code comments and strings as non-executable data, potentially allowing embedded instructions to be followed.
Capability inventory: The agent has the capability to execute bun test and bun run typecheck which runs code within the project, and it can perform file system operations via git worktree and git commit.
Sanitization: There is no evidence of sanitization, filtering, or escaping of the ingested code content before it is processed by the agent.
[COMMAND_EXECUTION]: The skill provides instructions for the agent to execute multiple shell commands to manage the development environment and validate simplification passes.
Evidence: SKILL.md and references/periodic-monorepo-pass.md include commands such as git fetch, git worktree add, bun test, bun run typecheck, rg, and fd.
[EXTERNAL_DOWNLOADS]: The skill fetches data from remote sources to synchronize with pull requests and to research library documentation.
Evidence: The skill uses git fetch to retrieve pull request data from GitHub and utilizes mcp__deepwiki__ask_question to query documentation for well-known external libraries like arktype, yjs, svelte, and hono.

collapse-pass