fluent-sdk

Pass

Audited by Gen Agent Trust Hub on Apr 29, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill presents an indirect prompt injection vulnerability surface because it ingests untrusted data from live ServiceNow instances. This data is then used to influence sensitive operations.
  • Ingestion points: Untrusted data enters the agent context via the SN-Query-Table MCP tool during verification and hybrid workflow steps (e.g., in SKILL.md Step 6).
  • Boundary markers: No specific delimiters or instructions to ignore embedded commands within the record data are provided to the agent.
  • Capability inventory: The skill possesses high-privilege capabilities including arbitrary server-side script execution via SN-Execute-Background-Script, fix script creation via SN-Create-Fix-Script, and code deployment via now-sdk deploy.
  • Sanitization: There is no evidence of sanitization or validation of the retrieved record content before it is interpolated into subsequent tool calls or script generation logic.
  • [EXTERNAL_DOWNLOADS]: The skill and its referenced vendor files (vendor/now-sdk-explain.md, vendor/now-sdk-setup.md) download the @servicenow/sdk package from the official NPM registry and reference source code from ServiceNow's official GitHub repository. These downloads originate from well-known services and trusted organizations.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 29, 2026, 04:36 PM