Public documentation for governed AI labor
SDKs/Governance/Connectors
Arx / Docs / security-ic-production-policy-drafter

Documentation

security-ic-production-policy-drafter

arxsec-site / reference-agents/security-ic-production-policy-drafter/README.md

arxsec-site reference-agents reference-agents/security-ic-production-policy-drafter/README.md

> Drafts security policies and control documentation from control frameworks and current state assessments. CISO reviews and approves.

Cell: security.ic · Shape: production · Source: arx-reference

Reference agent conforming to MANIFEST_FRAMEWORK.md. Apache 2.0 — fork, modify, or replace; the governance layer underneath stays.

What it does

Given a list of control domain IDs:

  1. Pulls supporting context through ARX-governed connectors
  2. Applies the security policy drafting logic
  3. Returns a structured policy draft for human review (or

auto-action if within declared low-risk scope)

  1. Streams every action to the customer's hash-chained personnel record

Governance posture

  • Onboarding — credentials issued at runtime, scoped per-call
  • Supervision — writes to github, confluence require manager approval
  • Evaluation — drift detection against manifests/job_description.yaml
  • Records — every action to customer S3 personnel-record bucket
  • Termination — one-button revoke + halt + exit attestation

Endpoints

| Method | Path | Description | | ------ | ----------- | ----------- | | POST | /act | Plan a security policy drafting pass against control domain IDs | | POST | /execute | Execute the plan | | GET | /status | Liveness + manifest sha256 + cell + shape |

Run locally

``bash docker build -t security-ic-production-policy-drafter:dev . docker run --rm -p 8100:8100 security-ic-production-policy-drafter:dev curl -s http://localhost:8100/status | jq ``