The Agent Skills Directory

[COMMAND_EXECUTION]: The skill performs extensive shell command execution for Git lifecycle management, including repository verification, branch creation, pulling updates, committing changes with specific co-author metadata, and pushing to remote origins. It also utilizes the GitHub CLI (gh) to automate the creation of Pull Requests with dynamically generated titles and descriptions.
[COMMAND_EXECUTION]: In the validation and testing phases (Steps 5 and 9), the skill dynamically executes build and test scripts based on the project environment. It recognizes and runs commands for multiple ecosystems, including Node.js (npm, yarn, pnpm), Rust (cargo), Go (go), and Python (ruff, mypy, pytest).
[PROMPT_INJECTION]: The skill exhibits an inherent surface for indirect prompt injection due to its core functionality of reading and analyzing untrusted codebase content.
Ingestion points: The workflow ingests data from local project configuration files (package.json, Cargo.toml, pyproject.toml, etc.) and performs a broad codebase analysis using a subagent in the SKILL.md (Step 2) and references/workflow.md files.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are used when interpolating codebase content or feature descriptions into subagent prompts.
Capability inventory: The skill possesses high-privilege capabilities within the execution environment, including full file system read/write access, network access via Git/GitHub CLI, and arbitrary command execution.
Sanitization: The skill does not implement sanitization or validation of the content read from the repository before it is used to influence the agent's planning or execution steps.

apex