Agent OS · Component · content-quality-judging
LLM editorial and outreach quality judging
What it does
Score draft and outreach quality, choose pass/rewrite/fail and manage judge-budget degradation.
Core pattern
Source and locale aware scoring + revision guidance
- Input
- Draft plus source and locale context
- Output
- Structured scores, verdict and rewrite guidance
Key parameters
| Name | Type | Default | Values |
|---|---|---|---|
| provider | string | — | — |
| model | string | — | — |
| temperature | number | 0 | 0–2 |
| max_output_tokens | integer | 4096 | 1–2000000 |
| budget_exhausted_policy | string | defer | fail_closed · defer · mark_unjudged |
Maturity and limits
Source paths were inspected statically; tests and live integrations were not executed. This rating describes implemented code, not production validation.
No root license found; platform access and automation terms apply. Credentials and browser sessions require isolation; public actions need explicit customer authorization.
1 sentence citing source paths or vendor names omitted for publication.
Used in pipelines
Not part of a pipeline yet; available as a standalone component.
Industry tags
- Government & public servicesTransferable use
For reviewing and publishing service guides and public explanations, use source and locale aware scoring followed by revision guidance.
Skills pack
Contract scaffold and deterministic mock
Download · 13KBSHA-256 602157c06b26ba2d5da4608207316e1cb115a885f9040b70c494261d175a09bf
Related components
-
Local coding-agent execution governance
Pre-action policy decisions + human permission and audit
Working 1 pipeline -
Collaborative delivery, approvals and evidence gates
Draft and artifact versioning + human handoff approval
Working 4 pipelines -
Tenant identity, enterprise SSO, provisioning and access control
Organization membership sync + scoped role access
Working 1 pipeline