deploy-to-connect
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONCREDENTIALS_UNSAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill manages the installation of the
rsconnect-pythonCLI and the Rrsconnectpackage. These are fetched from standard registries (PyPI) and official vendor infrastructure (Posit Package Manager), which is consistent with the skill's primary purpose. - [COMMAND_EXECUTION]: Uses system commands to detect project frameworks (e.g.,
grepfor imports inapp.py) and inventory capabilities (e.g.,command -v). These operations are restricted to environment discovery and deployment logic. - [CREDENTIALS_UNSAFE]: Contains proactive security directives that forbid the agent from scanning the environment,
.envfiles, or keychains for API keys. It enforces a browser-based OAuth flow for registering new servers, ensuring sensitive credentials are not handled directly by the agent. - [INDIRECT_PROMPT_INJECTION]: The skill ingests and analyzes local project files to infer deployment settings. While this represents a data ingestion surface, the logic is limited to identification (grep) and does not involve executing the contents of those files within the agent's context.
Audit Metadata