The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill directs users to install the claude-flow CLI tool globally using npm (npm install -g claude-flow@alpha). This tool is a primary requirement for the skill's functionality and is hosted on a well-known service.
[COMMAND_EXECUTION]: The skill heavily utilizes shell commands via the claude-flow CLI to manage development sessions, execute test suites, perform Git operations (such as commits, branching, and stashing), and run performance profiling.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted project files and source code, creating a surface for indirect prompt injection where malicious instructions hidden in the code could attempt to influence the AI's behavior.
Ingestion points: The agent ingests local source code files, configuration files, and git history during review, implementation, and refactoring tasks.
Boundary markers: There are no specific delimiters or 'ignore embedded instructions' warnings mentioned in the instructions to prevent the agent from obeying instructions found within the project files.
Capability inventory: The skill possesses extensive capabilities including file writing (/implement), code modification (/refactor), command execution for testing (/test, /perf), and version control operations (/commit).
Sanitization: The documentation does not describe any specific sanitization, validation, or escaping of the code content before it is interpolated into the agent's context.

Pair Programming