Skip to content

Security: Enforce consent and limits for external AI calls #9

@DerJanniku

Description

@DerJanniku

Goal

Define security boundaries for external AI providers and local models.

Scope

  • Add user consent gate before external data egress
  • Define allowlist/scoping for context payloads
  • Add budget/rate limits for agent-triggered actions

Acceptance Criteria

  • External provider calls require explicit opt-in
  • Limits are configurable and enforced
  • Security behavior documented in SECURITY.md

Metadata

Metadata

Assignees

No one assigned

    Labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions