The Agent Skills Directory

[COMMAND_EXECUTION]: The skill frequently executes shell commands using the Bash tool to interact with the GitHub CLI (gh) and run orchestration scripts.
[EXTERNAL_DOWNLOADS]: The skill utilizes npx ruv-swarm to dynamically download and execute the code review suite from the npm registry at runtime. This package aligns with the author's (ruvnet) namespace.
[INDIRECT_PROMPT_INJECTION]: The skill contains an indirect prompt injection surface by processing untrusted data from pull requests.
Ingestion points: External data enters the agent context via gh pr view and gh pr diff commands within SKILL.md to extract PR metadata and source diffs.
Boundary markers: There are no explicit boundaries, delimiters, or guidelines instructing the model to ignore instructions embedded within the pull request content.
Capability inventory: The system maintains full shell command execution via the Bash tool and file system modification capabilities via Read, Write, and TodoWrite tools.
Sanitization: Untrusted pull request text is directly assigned to shell variables ($PR_DATA, $PR_DIFF) and forwarded into the multi-agent system without prior escaping or structural validation.

agent-code-review-swarm