init-linux-cloud-agent-env
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Downloads the official GitHub CLI (
gh) binary from the official GitHub releases page and the Vercel CLI, both of which are managed by trusted organizations. - [EXTERNAL_DOWNLOADS]: Installs the Anthropic Python MCP CLI client, a library provided by a trusted AI research organization.
- [COMMAND_EXECUTION]: Executes shell commands to install, configure, and verify MCP servers and CLI tools on the Linux host.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process instructions from external GitHub repositories (specifically the 'superpowers' skill collection) and content from Pull Request descriptions.
- Ingestion points: External Markdown files from the 'superpowers' repository and project-specific documentation like
AGENTS.mdand PR bodies. - Boundary markers: No explicit delimiters are specified to separate external instruction content from the skill's core execution logic.
- Capability inventory: The agent has the ability to write to the filesystem, execute CLI tools (gh, vercel), and perform network operations via authorized MCP servers.
- Sanitization: The instructions do not define specific sanitization or validation steps for the external content before it is processed by the agent.
Audit Metadata