okta-cis-to-descope

Pass

Audited by Gen Agent Trust Hub on Jul 27, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses local shell commands such as grep and find to identify Okta implementation patterns and environment variables within the project codebase during the analysis phase.
  • [COMMAND_EXECUTION]: It performs network requests using curl to retrieve session, enrollment, and sign-on policies from the Okta Management API to inform the migration plan.
  • [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection as it processes data from the local codebase and external API responses.
  • Ingestion points: Reads local project source files and fetches remote policies from the Okta API.
  • Boundary markers: The instructions do not explicitly require the use of delimiters to isolate analyzed content from operational instructions.
  • Capability inventory: The agent can execute shell commands, perform file-write operations, and execute network calls.
  • Sanitization: There are no explicit requirements defined for sanitizing or filtering the data ingested from the project environment or external APIs.
  • [SAFE]: The skill handles the installation of official SDKs from standard package registries such as NPM and PyPI as part of the environment setup.
  • [SAFE]: The skill references and downloads configuration from trusted vendor domains, including official Descope and Okta resources.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 27, 2026, 08:24 AM
Security Audit — agent-trust-hub — okta-cis-to-descope