The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill recommends the use of npx @claude-flow/cli@latest. This operation downloads and executes the latest version of the @claude-flow/cli package from the NPM registry. The package name aligns with the mcp__claude-flow tools listed in the skill configuration, suggesting it is a component of the same workflow ecosystem.
[COMMAND_EXECUTION]: The skill relies on the Bash tool to perform local Git operations such as branching, merging, and rebasing, as well as executing the recommended CLI tool.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its core function of analyzing external, potentially untrusted data.
Ingestion points: The skill ingests repository data, health metrics, and code differences using mcp__claude-flow__github_repo_analyze, mcp__claude-flow__analyze_diff, and related tools (SKILL.md).
Boundary markers: There are no explicit boundary markers or instructions to the agent to disregard instructions embedded within the analyzed repository content or PR bodies.
Capability inventory: The agent has the capability to execute shell commands via Bash and modify repository states (e.g., managing PRs) via mcp__claude-flow__github_pr_manage (SKILL.md).
Sanitization: No sanitization or validation of the ingested repository content is mentioned in the skill instructions.

git-workflow