pp-gainsight
Pass
Audited by Gen Agent Trust Hub on Jul 28, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill downloads command-line binaries and libraries directly from the author's repositories on GitHub and NPM for the initial setup of the Gainsight CLI tool. (Evidence:
go install github.com/mvanhorn/printing-press-library/library/sales-and-crm/gainsight/cmd/gainsight-pp-cli@latestandnpx -y @mvanhorn/printing-press-library install gainsight). - [COMMAND_EXECUTION]: The skill instructions direct the agent to execute the
gainsight-pp-clibinary to perform various read-only operations on Gainsight data. (Evidence: Commands such asgainsight-pp-cli companies listandgainsight-pp-cli cockpit). - [DATA_EXFILTRATION]: The CLI includes a documented capability to deliver its output to a remote URL via the
--deliver webhook:<url>flag. (Evidence: The 'Output Delivery' section details how output can be POSTed to an external URL using thewebhookscheme). - [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection via its natural language command resolution feature. (Ingestion points: Processed user queries passed to the
gainsight-pp-cli whichcommand. Boundary markers: None present. Capability inventory: Network access via API and webhooks, and local file writing. Sanitization: Not explicitly described).
Audit Metadata