github-project-management
Warn
Audited by Gen Agent Trust Hub on May 10, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill frequently fetches and runs packages from the NPM registry using npx, specifically
claude-flow@alphaandruv-swarm. - [REMOTE_CODE_EXECUTION]: Remote code is executed directly via npx to initialize swarms, sync boards, and process issue data. This allows code from external repositories to run in the local environment.
- [COMMAND_EXECUTION]: Extensive use of shell scripts to interact with the GitHub CLI (gh) and parse project data using jq.
- [PROMPT_INJECTION]: The skill facilitates indirect prompt injection by acting on untrusted data from GitHub. Ingestion points: The skill reads issue titles, bodies, and comments using
gh issue viewandgh issue list. Boundary markers: No explicit markers or delimiters are used to wrap ingested issue content to prevent it from being interpreted as instructions. Capability inventory: The skill possesses the ability to write to GitHub (comments, edits), move project cards, and execute complex swarm logic. Sanitization: There is no evidence of sanitization or validation of the text ingested from issues before it is used to drive agent actions.
Audit Metadata