| Name | Modified | Size | Downloads / Week |
|---|---|---|---|
| Parent folder | |||
| README.md | 2026-04-13 | 1.1 kB | |
| v0.10.0 source code.tar.gz | 2026-04-13 | 10.3 MB | |
| v0.10.0 source code.zip | 2026-04-13 | 10.5 MB | |
| Totals: 3 Items | 20.8 MB | 0 | |
Added
- Usage metering hook —
RecursiveConfig.usage_callback: (RequestUsage, model_id) -> Nonefires once per pydantic-ai model request (orchestrator turns, sub-agent runs, tool-call follow-ups). Implemented viaace.rr.MeteredModel, apydantic_ai.models.wrapper.WrapperModelsubclass, so metering lives at the framework's own model boundary — one firing site, no per-call-site plumbing. Callback exceptions are caught and logged so metering never crashes the pipeline. - Pre-built model instance support —
RRStep,create_rr_agent,create_sub_agent, andRecursiveConfig.subagent_modelnow accept either a model-id string or a pre-builtpydantic_ai.models.Modelinstance. Enables callers that need a custom provider (e.g. a Bedrock model carrying STS-assumed credentials) to inject a fully-configured model rather than resolving from a string. - Sub-agent
model_settings—create_sub_agentnow threads an explicitModelSettingsparameter into itsPydanticAgentconstructor.
Back-compat
Existing RRStep(model="...") callers are unchanged. The widened type signatures are additive.