Current issue state, recent activity, and per-issue timelines from the indexed issue data.
| Date | Opened | Closed | Comments | Events | Open Backlog |
|---|---|---|---|---|---|
| 2026-09-07 | 0 | 0 | 1 | 1 | 22 |
| 2026-09-06 | 5 | 2 | 0 | 2 | 22 |
| 2026-09-05 | 5 | 1 | 0 | 0 | 39 |
| 2026-09-04 | 15 | 10 | 4 | 4 | 29 |
| 2026-09-03 | 13 | 9 | 0 | 0 | 0 |
| 2026-09-02 | 16 | 12 | 27 | 113 | 568 |
| 2026-09-01 | 7 | 7 | 0 | 0 | 0 |
| 2026-08-31 | 19 | 10 | 0 | 0 | 0 |
| 2026-08-30 | 11 | 2 | 0 | 0 | 0 |
| 2026-08-29 | 5 | 2 | 0 | 0 | 0 |
| 2026-08-28 | 8 | 7 | 0 | 0 | 0 |
| 2026-08-27 | 16 | 9 | 0 | 0 | 0 |
| 2026-08-26 | 13 | 4 | 0 | 0 | 0 |
| 2026-08-25 | 7 | 5 | 0 | 0 | 0 |
Opened: 61
Closed: 41
Comments: 32
Events: 120
| Issue | Author | State | Labels | Comments | Reactions | Updated |
|---|---|---|---|---|---|---|
#5685 [aw] No-Op Runs Opened 3 months ago | github-actions[bot] | open | meta pydanty:meta agentic-workflows | 54 | 0 | 2 hours ago |
#8183 [aw] CI Review failed Opened 7 hours ago | github-actions[bot] | open | agentic-workflows | 2 | 0 | 2 hours ago |
#8191 Model equality ignores `settings`, so merged capabilities drop the later `fallback_subagent_model` Opened 4 hours ago | dsfaccini | open | bug | 1 | 0 | 2 hours ago |
#8193 [aw] CI Review hit engine rate limit (HTTP 429) Opened 3 hours ago | github-actions[bot] | open | agentic-workflows | 2 | 0 | 2 hours ago |
#8184 Reduce capability traceback depth and exception-local serialization overhead Opened 6 hours ago | dsfaccini | open | performance tracking pydanty:skip | 0 | 0 | 3 hours ago |
#8176 Refusal handling diverges across providers: OpenAI raises ContentFilterError, Anthropic/Google return the refusal as a normal result Opened 12 hours ago | ptimizeroracle | closed - not_planned | pydanty:bug p:4-low | 4 | 0 | 3 hours ago |
#8181 VercelAIAdapter.deferred_tool_results carries approvals only, never external call results Opened 7 hours ago | RemyOstyn | open | No labels | 0 | 0 | 4 hours ago |
#8182 Profile of the full `pyright` run and what would shrink it Opened 7 hours ago | dsfaccini | open | No labels | 1 | 0 | 5 hours ago |
#8186 Refactor `tests/test_capability...` into `tests/capabilities/test_...` + set a token guard to keep test files below 10k tokens Opened 6 hours ago | dsfaccini | open | meta chore test-suite | 0 | 0 | 6 hours ago |
#8185 Vercel AI stream surface has no carrier for tool outcome claims — echoed turns reload `retried` as `failed` Opened 6 hours ago | dsfaccini | open | No labels | 0 | 0 | 6 hours ago |
#6927 `run_on_errors` evaluators lose captured spans and retain exporter state after task exceptions Opened 1 month ago | cosin2077 | open | needs-maintainer-action | 2 | 0 | 6 hours ago |
#6932 Add audio content support when mapping Pydantic AI messages to MCP sampling messages Opened 1 month ago | emredeveloper | open | needs-maintainer-action | 1 | 0 | 6 hours ago |
#5687 Add per-turn index to GenAI span names for agent/model requests Opened 3 months ago | sipa-echo-ngbm | open | feature OpenTelemetry evals | 11 | 0 | 6 hours ago |
#7900 Capability hook for turn-scoped state declaration — volatile state in the history tail instead of per-request head instructions Opened 9 days ago | jimmyclaw1994-pixel | open | performance pydanty:feature p:3-mid | 3 | 0 | 6 hours ago |
#7975 OTel usage attributes: should subset detail keys (reasoning/thinking/cache) be marked or protected against sum-consumers? Opened 7 days ago | roy-tong | open | question pydanty:feature documentation | 4 | 0 | 7 hours ago |
#3970 Track `ty` type checker readiness for pydantic-ai Opened 8 months ago | dsfaccini | open | meta chore tracking | 4 | 9 | 7 hours ago |
#530 Ability to Persist Messages in External Stores Opened 2 years ago | izzyacademy | open | feature community snippets message-history | 28 | 0 | 9 hours ago |
#8045 OpenAI Responses streaming retains the requested router alias instead of the terminal served model Opened 4 days ago | MTG-Thomas | open | pydanty:bug p:3-mid | 5 | 0 | 10 hours ago |
#7888 a request Opened 9 days ago | Abdulmohsen-almutlaq | open | question Stale | 4 | 0 | 11 hours ago |
#8127 Two capabilities that both need the innermost toolset slot: an authorize-and-record toolset and a durability engine Opened 3 days ago | rafaelasor | open | pydanty:feature p:4-low | 7 | 0 | 11 hours ago |
#6460 Cancellation semantics: a level-triggered contract for agent runs, streams, tools, and durable execution Opened 2 months ago | DouweM | open | feature graph asyncio | 7 | 0 | 12 hours ago |
#7785 `join()` with a fanned producer silently drops that branch's contribution Opened 12 days ago | FIITAY | open | bug pydanty:bug p:2-high | 4 | 1 | 12 hours ago |
#8018 CI Review agent job fails to check out the PR head: pre-warm rewrites pydantic-ai-runner.lock Opened 5 days ago | hassaanch23 | open | needs-maintainer-action | 1 | 0 | 12 hours ago |
#7770 `CallDeferred` raised from a model-request hook escapes the run with an empty message Opened 13 days ago | dsfaccini | open | needs-maintainer-action | 2 | 0 | 12 hours ago |
#8130 `BedrockConverseModel` lets botocore transport errors (`ReadTimeoutError`, `EndpointConnectionError`) escape unwrapped, so `FallbackModel` does not fall back on them Opened 3 days ago | kimnamu | closed - completed | bedrock p:2-high pydanty:pr | 1 | 0 | 14 hours ago |