implement-saaskit-python

Pass

Audited by Gen Agent Trust Hub on Aug 28, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill implements an authentication flow that ingests untrusted data from external sources.
  • Ingestion points: The skill processes authentication data via OAuth callback parameters (code, state) and session management through the sk_session cookie across SKILL.md, references/django.md, and references/flask.md.
  • Boundary markers: The skill relies on the scalekit-sdk-python library to manage and validate these inputs.
  • Capability inventory: The agent's task involves creating and modifying application source code (app.py, settings.py, urls.py, views.py).
  • Sanitization: Input validation and sanitization (such as for the returnTo redirect parameter) are delegated to the vendor-provided adapter as noted in SKILL.md.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 28, 2026, 08:07 AM
Security Audit — agent-trust-hub — implement-saaskit-python