github-actions
Pass
Audited by Gen Agent Trust Hub on May 1, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Documents the installation of the @membranehq/cli package from the public NPM registry. This tool is the official CLI for the skill's associated service.
- [COMMAND_EXECUTION]: Utilizes a dedicated CLI tool to interact with GitHub Actions, enabling tasks such as listing workflows, searching for actions, and executing jobs.
- [PROMPT_INJECTION]: The skill ingests untrusted data from GitHub Actions (e.g., workflow run data, job outputs) and possesses the capability to execute actions via the CLI. This surface for indirect prompt injection is noted as an inherent part of the integration's functionality.
- [SAFE]: Implements secure credential management by leveraging the Membrane platform for authentication, explicitly instructing the agent to avoid direct requests for user secrets or API keys.
Audit Metadata