All pull requests

Pull requests

• fix(ai-cloudflare): split embed() inputs into batches of at most 100

  #2036·pcbeingused333 opened Nov 4, 2026Contributor·4 of 7 tasks·

  ·
  4

• feat(ai-sandbox): stream sandbox lifecycle events; fix gateway image model type-check

  #2033·AlemTuzlak opened Nov 4, 2026Contributor·5 of 7 tasks·

  ·
  9

• fix(ai-openai): split embed() inputs into requests of at most 2048

  waiting-on: maintainer
  #1986·pcbeingused333 opened Nov 4, 2026Contributor·4 of 7 tasks·

  ·
  6

  

• feat(ai-computer-sandbox, ai-sandbox-docker, ai-sandbox-cloudflare, examples): sign in on a sandbox desktop without the model seeing the password, and take over when it is stuck

  merge-conflictswaiting-on: author
  #1982·tombeckenham opened Nov 3, 2026Contributor·5 of 7 tasks·

  ·
  41

  

• ci: use voidzero-dev/setup-vp directly

  waiting-on: author
  #1980·lachlancollins opened Nov 2, 2026Member·4 of 7 tasks·

  ·
  10

  

• fix(ai-client): show server-only tool approvals as server-tool-approval

  waiting-on: maintainer
  #1967·tombeckenham opened Nov 2, 2026Contributor·5 of 7 tasks·

  ·
  1

  10

  

• feat(ai): live speech-to-text sessions with Deepgram, ElevenLabs, OpenAI, xAI, Gemini, Mistral, and Cloudflare adapters

  waiting-on: maintainer
  #1964·tombeckenham opened Nov 1, 2026Contributor·5 of 7 tasks·

  ·
  1

  21

  

• fix(ai-client): coalesce the per-chunk snapshot storm while streaming

  #1956·jherr opened Oct 30, 2026Contributor·6 of 7 tasks·

  ·
  5

• fix(ai-harness): store a rejected input with its rejection

  waiting-on: maintainer
  #1955·tombeckenham opened Oct 30, 2026Contributor·5 of 7 tasks·

  ·
  1

  8

  

• feat(ai-sandbox-render): add Render sandbox provider

  merge-conflictswaiting-on: author
  #1946·R4ph-t opened Oct 29, 2026·6 of 7 tasks·

  ·
  18

  

• fix(ai-client): keep the replay's answer when a rejoin starts with reasoning

  waiting-on: maintainer
  #1942·aayushbtw opened Oct 29, 2026Contributor·6 of 7 tasks·

  ·
  1

  3

  

• feat(ai-zai): add Z.AI adapter on the OpenAI chat completions base

  waiting-on: maintainer
  #1937·shakibdshy opened Oct 28, 2026·6 of 7 tasks·

  ·
  24

  

• feat(ai-client): send AG-UI protocolVersion in the run request body

  waiting-on: maintainer
  #1934·kbuschgens opened Oct 28, 2026·6 of 7 tasks·

  ·
  3

  

• fix(openai-base): finish a Responses max_output_tokens cutoff with finishReason length instead of RUN_ERROR

  waiting-on: maintainer
  #1929·pcbeingused333 opened Oct 28, 2026Contributor·3 of 7 tasks·

  ·
  1

  5

  

• feat(ai-generative-ui): Open Generative UI — Tier F artifacts, MCP Apps, live data, and A2UI inbound surfaces

  waiting-on: maintainer
  #1927·jherr opened Oct 27, 2026Contributor·5 of 7 tasks·

  ·
  21

  

• fix(ai-ollama): report finishReason length for truncated Ollama streams

  waiting-on: author
  #1902·TPSJASPER opened Oct 25, 2026Contributor·5 of 7 tasks·

  ·
  1

  6

  

• feat(ai-harness): session.seed starts an empty thread with existing messages

  waiting-on: maintainer
  #1898·AlexDemzz opened Oct 23, 2026Contributor·5 of 7 tasks·

  ·
  25

  

• feat(examples/ts-react-chat): Open Generative UI Tier F demo with live data

  merge-conflictswaiting-on: author
  #1896·jherr opened Oct 23, 2026Contributor·6 of 7 tasks·

  ·
  6

  

• feat(ai-ui-message-stream): bridge TanStack AI and the AI SDK UI message stream

  waiting-on: maintainer
  #1893·tombeckenham opened Oct 23, 2026Contributor·5 of 7 tasks·

  ·
  13

  

• feat(ai-models): build the catalog from modelschemas, not models.dev

  waiting-on: author
  #1791·tombeckenham opened Oct 15, 2026Contributor·5 of 7 tasks·

  ·
  1

  10

  

• fix(ai, ai-anthropic): skip summarize temperature on no-sampling models

  ai-readysecurewaiting-on: author
  #1705·harshlocham opened Oct 10, 2026Contributor·8 of 9 tasks·

  ·
  1

  11

  

• fix(ai): propagate errors from onAfterToolCall

  waiting-on: maintainer
  #1559·dakjdakd opened Sep 28, 2026Contributor·5 of 7 tasks·

  ·
  1

  13

  

• feat(model-sync): sync native models from modelschemas only, with no prices in model metadata

  merge-conflictswaiting-on: author
  #1308·tombeckenham opened Sep 3, 2026Contributor·5 of 7 tasks·

  ·
  23

  

• Add Workflow-backed AI orchestration

  #979·tannerlinsley opened Jul 21, 2026Member·

  ·
  12