generated: '2026-08-06'
method: generated
source: openapi/activefence-alice-api-openapi.yml
provider: ActiveFence (Alice)
note: >-
  Alice publishes no AGENTS.md and no skill files of its own — the GitHub org (3 public repos)
  and the docs host were searched. These are generated from the provider's own OpenAPI, with
  every operationId verified verbatim against the harvested spec, plus the conventions, error
  and webhook rules captured elsewhere in this repo.
skills:
- file: activefence-moderate-user-content.md
  name: Moderate user-generated content with ActiveScore
  description: >-
    Submit text, image, video or audio for risk scoring; choose the synchronous pre-publish gate
    or the asynchronous callback path; interpret risk_score and violation_type keys.
  api: openapi/activefence-alice-api-openapi.yml
  product: ActiveFamily
  operations:
  - post-sync-content-text
  - post-sync-content-bulk-text
  - post-sync-content-image
  - post-content-text
  - post-content-image
  - post-content-video
  - post-content-audio
- file: activefence-guard-genai-messages.md
  name: Guard GenAI prompts and responses with WonderFence
  description: >-
    Place a runtime guardrail on both sides of an LLM call to catch prompt injection,
    jailbreaks, PII exposure and denied topics, with session-scoped multi-turn detection.
  api: openapi/activefence-alice-api-openapi.yml
  product: WonderFence
  operations:
  - post-genai-evaluate
- file: activefence-run-redteam-assessment.md
  name: Run a WonderBuild red-team assessment
  description: >-
    Drive pre-deployment adversarial testing from CI — launch from a preset, clone-and-run for
    regression, and poll the assessment status lifecycle as a release gate.
  api: openapi/activefence-alice-api-openapi.yml
  product: WonderBuild
  operations:
  - get-assessments
  - run-assessment
  - clone-and-run-assessment
  - run-assessment-preset