skills/celigo/ai/writing-scripts/Gen Agent Trust Hub

writing-scripts

Pass

Audited by Gen Agent Trust Hub on Jul 23, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The documentation provides examples of CLI commands (e.g., celigo scripts get, celigo flows run) used to interact with the Celigo platform. These commands are intended for developers to manage and debug integrations within the platform's ecosystem.
  • [PROMPT_INJECTION]: The skill was evaluated for potential indirect injection surfaces (Category 8).
  • Ingestion points: Data pipeline hooks such as preSavePage and preMap ingest records directly from external data sources into the script context.
  • Boundary markers: The documentation does not define specific boundary markers for external data; objects are processed as standard JavaScript inputs.
  • Capability inventory: Scripts have access to the integrator-api module for outbound operations and sjcl for cryptographic functions.
  • Sanitization: Sanitization practices are left to the implementation of the script author, as is standard for programmable integration platforms.
  • [SAFE]: The skill serves as official technical documentation for the Celigo platform. All referenced tools, libraries (such as dayjs and sjcl), and internal modules are appropriate and expected components for the documented SaaS functionality.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 23, 2026, 06:09 AM
Security Audit — agent-trust-hub — writing-scripts