The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill documentation includes an alternative command npx @claude-flow/cli@latest which triggers a download of an external package from the npm registry at runtime.
[REMOTE_CODE_EXECUTION]: Executing an unverified package via npx results in the execution of code fetched from a remote source. The package @claude-flow/cli is not associated with a trusted organization or the skill author's verified naming patterns.
[COMMAND_EXECUTION]: The skill uses Bash to perform git operations and execute the npx utility in the provided CLI example.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from external repositories and git diffs, making it vulnerable to instructions hidden within the data being analyzed.
Ingestion points: Untrusted data enters the agent context via mcp__plugin_ruflo-core_ruflo__github_repo_analyze and mcp__plugin_ruflo-core_ruflo__analyze_diff as specified in SKILL.md.
Boundary markers: The instructions lack delimiters or explicit directives to ignore embedded instructions within the ingested git content.
Capability inventory: The skill possesses capabilities including Bash execution and GitHub Pull Request management (mcp__plugin_ruflo-core_ruflo__github_pr_manage).
Sanitization: No validation or sanitization of the repository data or diff content is performed before processing.

git-workflow