The Agent Skills Directory

[COMMAND_EXECUTION]: The skill facilitates the execution of various development and system commands. It uses tools like git for version control, gh for managing issues and pull requests, and package managers like uv and bun to run linters, type checkers, and tests. It also includes specific instructions for backgrounding long-running processes using nohup and disown to ensure stability.
[PROMPT_INJECTION]: The skill is designed to ingest and process data from external sources such as GitHub issues (gh issue view) and pull request checks, which creates a surface for indirect prompt injection.
Ingestion points: External text is ingested from GitHub issues and pull request statuses as described in the workflow steps in SKILL.md.
Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded commands within the ingested data.
Capability inventory: The skill has capabilities to modify the file system, execute shell commands (via uv, bun, git, and gh), and interact with the remote repository.
Sanitization: No explicit sanitization or filtering of the content retrieved from GitHub is specified.

engineering-loop