CircleAI.ContentPolicy
(3.0.1) Content-policy / refusal contracts — pattern-adoption of Sponsio. Distinct from the v1.2.0 CircleAI.Safety personal-safety domain pack. IContentFilter (pre-completion guardrails) + IRefusalPolicy + IPromptInjectionDetector (second-order attacks via RAG/web/tool output) + ISafetyAuditLog. Null impls fail-closed (treat everything as refused, log nothing). Real backends in 3.0.2. Renamed from CircleAI.Guardrails 3.0.0 — that package ID exists on nuget.org but is a dead-end; nothing newer will ship under that ID. New consumers: `dotnet add package CircleAI.ContentPolicy`.
View on Github
Activity
- Latest release
- 2mo ago
- Total releases
- 3
- Cadence
- ~2 days
- Last 12 months
- 3
Reach
- Stars
- —
Details
- License
- MIT
- First release
- Jun 17, 2026
Releases
| Version | Released | |
|---|---|---|
3.2.0
minor
| ||
3.1.0
minor
| ||
3.0.1
initial
|