Skip to content

[Bug]: Antigravity is not working after recent changes #12206

Closed
#13389
#2829
Closed
[Bug]: Antigravity is not working after recent changes#12206
#13389
#2829
Labels
bugSomething is broken or behaving incorrectly.needs more infoInitial triage showed no bug. Awaiting more infovia-triageFiled through npx t3 triage

Description

@pkreddy8223
pkreddy8223
opened on Sep 17, 2026

Before submitting

  • I searched existing issues and did not find a duplicate.
  • I included enough detail to reproduce or investigate the problem.

Area

apps/web

Steps to reproduce

Antigravity is not working

Expected behavior

Antigravity is not working

Actual behavior

Antigravity is not working

Impact

Blocks work completely

Version or commit

No response

Environment

Win 11

Logs or stack traces

ProviderAdapterSessionClosedError: antigravity adapter thread is closed: c010e15b-01d8-4bfc-95d1-5880dd2e2f8e at mapAcpToAdapterError (file:///C:/Users/Pavan/AppData/Local/Programs/t3code/resources/server.asar/apps/server/dist/bin.mjs:146178:45) at mapAntigravityError (file:///C:/Users/Pavan/AppData/Local/Programs/t3code/resources/server.asar/apps/server/dist/bin.mjs:159801:7) at file:///C:/Users/Pavan/AppData/Local/Programs/t3code/resources/server.asar/apps/server/dist/bin.mjs:160358:201 at file:///C:/Users/Pavan/AppData/Local/Programs/t3code/resources/server.asar/apps/server/dist/claudeHistoryWorker-CAn1PawV.mjs:10021:96 at ~effect/Utils/internal (file:///C:/Users/Pavan/AppData/Local/Programs/t3code/resources/server.asar/apps/server/dist/claudeHistoryWorker-CAn1PawV.mjs:2558:10) at startSession (file:///C:/Users/Pavan/AppData/Local/Programs/t3code/resources/server.asar/apps/server/dist/bin.mjs:205107:59) at startSession (definition) (file:///C:/Users/Pavan/AppData/Local/Programs/t3code/resources/server.asar/apps/server/dist/bin.mjs:192276:23) at ensureSessionForThread (file:///C:/Users/Pavan/AppData/Local/Programs/t3code/resources/server.asar/apps/server/dist/bin.mjs:205191:10) at ensureSessionForThread (definition) (file:///C:/Users/Pavan/AppData/Local/Programs/t3code/resources/server.asar/apps/server/dist/bin.mjs:205026:33) at processTurnStartRequested (file:///C:/Users/Pavan/AppData/Local/Programs/t3code/resources/server.asar/apps/server/dist/bin.mjs:205752:12) at processTurnStartRequested (definition) (file:///C:/Users/Pavan/AppData/Local/Programs/t3code/resources/server.asar/apps/server/dist/bin.mjs:205405:36) at processDomainEvent (file:///C:/Users/Pavan/AppData/Local/Programs/t3code/resources/server.asar/apps/server/dist/bin.mjs:205780:46) at processDomainEvent (definition) (file:///C:/Users/Pavan/AppData/Local/Programs/t3code/resources/server.asar/apps/server/dist/bin.mjs:205729:29) { [cause]: AcpProcessExitedError: ACP process exited with code 4294967295 at Object.onSuccess (file:///C:/Users/Pavan/AppData/Local/Programs/t3code/resources/server.asar/apps/server/dist/bin.mjs:109331:23) }

Hi

10:37 am


ProviderAdapterSessionClosedError: antigravity adapter thread is closed: c010e15b-01d8-4bfc-95d1-5880dd2e2f8e
    at mapAcpToAdapterError (file:///C:/Users/Pavan/AppData/Local/Programs/t3code/resources/server.asar/apps/server/dist/bin.mjs:146178:45)
    at mapAntigravityError (file:///C:/Users/Pavan/AppData/Local/Programs/t3code/resources/server.asar/apps/server/dist/bin.mjs:159801:7)
    at file:///C:/Users/Pavan/AppData/Local/Programs/t3code/resources/server.asar/apps/server/dist/bin.mjs:160358:201
    at file:///C:/Users/Pavan/AppData/Local/Programs/t3code/resources/server.asar/apps/server/dist/claudeHistoryWorker-CAn1PawV.mjs:10021:96
    at ~effect/Utils/internal (file:///C:/Users/Pavan/AppData/Local/Programs/t3code/resources/server.asar/apps/server/dist/claudeHistoryWorker-CAn1PawV.mjs:2558:10)
    at startSession (file:///C:/Users/Pavan/AppData/Local/Programs/t3code/resources/server.asar/apps/server/dist/bin.mjs:205107:59)
    at startSession (definition) (file:///C:/Users/Pavan/AppData/Local/Programs/t3code/resources/server.asar/apps/server/dist/bin.mjs:192276:23)
    at ensureSessionForThread (file:///C:/Users/Pavan/AppData/Local/Programs/t3code/resources/server.asar/apps/server/dist/bin.mjs:205191:10)
    at ensureSessionForThread (definition) (file:///C:/Users/Pavan/AppData/Local/Programs/t3code/resources/server.asar/apps/server/dist/bin.mjs:205026:33)
    at processTurnStartRequested (file:///C:/Users/Pavan/AppData/Local/Programs/t3code/resources/server.asar/apps/server/dist/bin.mjs:205752:12)
    at processTurnStartRequested (definition) (file:///C:/Users/Pavan/AppData/Local/Programs/t3code/resources/server.asar/apps/server/dist/bin.mjs:205405:36)
    at processDomainEvent (file:///C:/Users/Pavan/AppData/Local/Programs/t3code/resources/server.asar/apps/server/dist/bin.mjs:205780:46)
    at processDomainEvent (definition) (file:///C:/Users/Pavan/AppData/Local/Programs/t3code/resources/server.asar/apps/server/dist/bin.mjs:205729:29) {
  [cause]: AcpProcessExitedError: ACP process exited with code 4294967295
      at Object.onSuccess (file:///C:/Users/Pavan/AppData/Local/Programs/t3code/resources/server.asar/apps/server/dist/bin.mjs:109331:23)
}

Screenshots, recordings, or supporting files

Screenshot 2026-09-17 104018.png

Workaround

No response

Activity

  1. pkreddy8223
    added
    bugSomething is broken or behaving incorrectly.
    needs-triageIssue needs maintainer review and initial categorization.
    on Sep 17, 2026

  2. juliusmarminge commented on Sep 17, 2026

    @juliusmarminge
    juliusmarminge
    on Sep 17, 2026
    Member

    Triage

    This is a Windows desktop Antigravity session-start crash, not a web UI bug. The chat error is T3 wrapping a dead ACP child.

    ProviderAdapterSessionClosedError is produced by mapAcpToAdapterError when the Antigravity process exits. The cause is:

    AcpProcessExitedError: ACP process exited with code 4294967295

    4294967295 is unsigned -1. On Windows that usually means agy_acp_server.exe crashed or was killed during spawn / session/start, not a clean ACP close.

    The stack is:

    processTurnStartRequested → ensureSessionForThread → startSession → mapAntigravityError

    in the packaged desktop server (…\t3code\resources\server.asar\…). The screenshot is the same stack on a local-checkout thread (Gemini 3.8 Flash High).

    Not a duplicate of the mid-turn reconnect issues

    This is not #11670 (dropped streamGenerateContent / WebSocket 1000 mid-turn). Open recovery PRs (#11719, #11705, #12092) do not explain an immediate -1 exit at session start.

    Also not #11414 (SIGILL / AVX) or #9800 (WSL IPv6 SIGABRT).

    Likely code

    • apps/server/src/provider/Layers/AntigravityAdapter.ts
    • apps/server/src/provider/acp/AcpAdapterSupport.ts
    • apps/server/src/provider/acp/AcpSessionRuntime.ts
    • apps/server/src/provider/Drivers/AntigravityDriver.ts
    • apps/server/src/provider/antigravityAuthSupport.ts (TEMP/TMP on Windows)

    Startup stderr is still discarded today (#9839), so the UI only shows this generic wrapper.

    Recent Windows change to check

    #12008 (fixes #9650) isolates Antigravity TEMP/TMP under the provider profile so PyInstaller _MEI* extracts stop filling %TEMP%. It merged 2026-09-16, is not in v0.0.42, and is in nightlies from 0.0.43-nightly.20260916.1811. That PR was not verified on a real Windows machine. If this build includes it, a bad isolated temp dir is the first hypothesis.

    If this is v0.0.42, #12008 is not present. Then leftover _MEI* folders / low disk from the old health-check leak, or a native crash of agy_acp_server.exe, are more likely.

    Need from you

    1. Exact T3 version — Settings / About, or the installer name (0.0.42 vs 0.0.43-nightly.YYYYMMDD.*).
    2. Did Antigravity work on the previous build, and did this start right after an update?
    3. Settings → Providers → Antigravity: installed? signed in? any status message?
    4. Free disk space, and whether %LOCALAPPDATA%\Temp has many large _MEI* folders.
    5. Logs (scrub account/project ids):
      • %USERPROFILE%\.t3\userdata\logs\server.trace.ndjson
      • provider event logs under %USERPROFILE%\.t3\userdata\logs\ if present

    Meanwhile

    • If %TEMP% is full of _MEI* folders: quit T3, delete those folders, free disk, retry. On 0.0.42 you can set Settings → Providers → Advanced → Provider health check interval to 0 so they stop coming back (#9650).
    • Try Refresh provider status, or reinstall the Antigravity runtime from provider settings.
    • If you are on a 2026-09-16+ nightly and this started after that update, say so — that points at #12008.

    Leaving this open as a Windows Antigravity session-start failure. Next step for maintainers: pin the version, then either reproduce #12008’s temp isolation on Windows or inspect native stderr / a crash dump from agy_acp_server.exe.

  3. juliusmarminge
    added
    via-triageFiled through npx t3 triage
    needs more infoInitial triage showed no bug. Awaiting more info
    and removed
    needs-triageIssue needs maintainer review and initial categorization.
    on Sep 17, 2026
  4. juliusmarminge
    closed this as completedin #13389on Sep 24, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    bugSomething is broken or behaving incorrectly.needs more infoInitial triage showed no bug. Awaiting more infovia-triageFiled through npx t3 triage

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions