The Agent Skills Directory

[COMMAND_EXECUTION]: The skill coordinates the execution of project-defined test suites (e.g., npm test, pytest) and building processes across multiple background agents. It also utilizes the GitHub CLI (gh) to push branches and submit pull requests.- [EXTERNAL_DOWNLOADS]: The skill interacts with GitHub for code management and pull request creation, which involves network communication with a well-known service.- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it reads and processes arbitrary codebase content and user-provided arguments to dynamically generate prompts for subagents without applying sanitization or clear boundary delimiters.
Ingestion points: Data enters the agent's context through user instructions ($ARGUMENTS) and the automated research of codebase files in Phase 1.
Boundary markers: The skill does not use explicit delimiters or instructions to isolate untrusted data from the core system prompt within the worker instructions.
Capability inventory: The skill possesses the capability to spawn background agents, execute shell commands (tests and git operations), and perform file system writes.
Sanitization: No sanitization, filtering, or validation is performed on the content retrieved from the codebase before it is interpolated into instructions.

batch