pydantic-ai · Issues· 921 open

[aw] No-Op Runs

OpenAI strict mode: `prefixItems` (tuple fields) is marked strict-compatible but is rejected; `minLength`/`maxLength` are marked incompatible but are accepted

`CombinedToolset.call_tool` dispatches through a stale cached tool, dropping `tool_def` mutations made by outer wrapper toolsets

Guard against discriminated-union / discriminator-tag drift in `messages.py` (test-suite robustness)

Stream and run cancellation semantics tracker

Document that capability hooks read the activity set as of request preparation, and that `prepare_tools` is the tool gate

Run the latest-versions canary on a nightly schedule, not only on tag refs

Usage a capability adds to `ctx.usage` inside a durable operation is credited on replay but not on a first run

Run an agent on a model that only answers questions: what `TypeSafeModel` (Jev) still needs from the framework

Server-initiated sampling and elicitation are unreachable on FastMCP 4

Separate effective tool definitions from tool execution state

`OnlineEvaluation` exports prompts via evaluation explanation logs regardless of `include_content`

`RequestUsage.extract` swallows all exceptions and returns a fabricated zero usage

`gpt-oss` resolves `supports_thinking=False` on heroku, nebius, ollama and ovhcloud — `harmony_model_profile` delegates reasoning to a table that excludes gpt-oss

Prompted output templates reject literal JSON braces