Repository navigation
Closed as duplicate
Closed as duplicate
Copy link
Description
Gemini 4 Neon on behalf of sol-tau
Before submitting
- I searched existing issues and did not find a duplicate. (Checked [Bug]: Antigravity ACP drops the 8 T3 MCP tools >1,280B (incl. delegate_task & t3_thread_launch) without call_mcp_tool or T3_ACP_MCP_NODE fallback #38123, fix(antigravity): defer optional T3 MCP tools via tool_call proxy #39362, [Bug]: schedule_task can't set model or reasoning effort, and silently ignores modelSelection #16833, [Bug]: Scheduled tasks freeze the creating thread's runtime mode, so runs ignore later changes to the default #32237, [Bug]: schedule_task on a linked environment fails when the project uses the environment's default model #37291. The underlying tool-eviction cause is [Bug]: Antigravity ACP drops the 8 T3 MCP tools >1,280B (incl. delegate_task & t3_thread_launch) without call_mcp_tool or T3_ACP_MCP_NODE fallback #38123, while this report covers the distinct failure mode where Antigravity's eagerly declared native
scheduletool hijacks scheduling turns—leaving T3 threads stuck inWorkingand broken across restarts—which can persist even whenschedule_taskis deferred behind a proxy as in fix(antigravity): defer optional T3 MCP tools via tool_call proxy #39362.) - I included enough detail to reproduce or investigate the problem.
Area
apps/server
Steps to reproduce
- Run T3 Code (
v0.1.12-nightly.20261122.5098) with the Antigravity ACP provider (agy 1.5.14, Google account,gemini-4-neon) on a profile with ~35 installed skills, so Antigravity's shared customization budget evicts the largestt3-codeMCP schemas (as in [Bug]: Antigravity ACP drops the 8 T3 MCP tools >1,280B (incl. delegate_task & t3_thread_launch) without call_mcp_tool or T3_ACP_MCP_NODE fallback #38123), includingschedule_task(3,318 B). - Ask the agent in a thread to set a one-shot reminder for ~8–9 hours later (e.g., "remind me tomorrow morning").
- Because
mcp_t3-code_schedule_taskis missing from native tool declarations (andcall_mcp_toolis absent), while Antigravity's built-inscheduletool (DurationSeconds/CronExpression) is eagerly declared, the agent invokes Antigravity's nativescheduletool. - Observe the thread state in T3 Code and then restart the T3 Code server.
Expected behavior
- Scheduling a persistent reminder or cron from an Antigravity thread should use T3 Code's
schedule_task(via direct MCP tool or theacp-mcp-call schedule_taskfallback) so the scheduled task persists across restarts and the thread immediately returns toIdle. <t3_code_instructions>for Antigravity should explicitly tell the model not to use its native in-sessionscheduletool for reminders/crons whenschedule_taskis absent or deferred, and instead callacp-mcp-call schedule_task(or surface the proxy tool cleanly once fix(antigravity): defer optional T3 MCP tools via tool_call proxy #39362 lands).
Actual behavior
- Antigravity's native
scheduletool spawns an in-session background timer inside the active turn. - T3 Code keeps the thread in
Workingstate indefinitely (e.g.,Working for 1h 51m 1swithRun schedulestill spinning in the UI) instead of returning toIdle. - When T3 Code restarts, the in-memory background timer is killed (
All your subagents and background tasks have been stopped due to server restart), and T3 Code printsT3 Code restarted and resumed this turn/ sendsContinue where you left off, which causes the agent to re-create another in-session timer if it doesn't switch toacp-mcp-call.
Impact
Minor bug or occasional failure
Version or commit
t3 v0.1.12-nightly.20261122.5098
Environment
Linux 7.2.15-1-omarchy (x64), Antigravity ACP 1.5.14 (agy 1.5.14, Google account, gemini-4-neon), 35 skills installed
Logs or stack traces
# Live session inspection of <appDataDir>/mcp/t3-code/*.json (86 schemas total)
# 80 schemas <= 1521 B loaded eagerly as mcp_t3-code_*
# 6 schemas >= 1629 B silently omitted (no Lazy: entries, no call_mcp_tool):
3318B DROPPED schedule_task
2915B DROPPED t3_thread_launch
1926B DROPPED delegate_task
1781B DROPPED t3_project_update
1668B DROPPED html_render
1629B DROPPED preview_act
1521B EAGER preview_open
# After Antigravity invoked native `schedule` (8h timer) and T3 server restarted:
# UI showed: "T3 Code restarted and resumed this turn" + "Working for 1h 51m 1s" + spinning "Run schedule"
# System notice delivered to agent:
[Notice] All your subagents and background tasks have been stopped due to server restart.
Workaround
Cancel the Antigravity native schedule background task (via manage_task), then create the persistent scheduled task using the documented acp-mcp-call terminal fallback:
ELECTRON_RUN_AS_NODE=1 "$T3_ACP_MCP_NODE" ${T3_ACP_MCP_ENTRYPOINT:+"$T3_ACP_MCP_ENTRYPOINT"} \
acp-mcp-call schedule_task '{"title":"...","prompt":"...","schedule":{"type":"once","runAt":"..."},"bindToCurrentThread":true,"enabled":true}'The thread immediately returns to Idle and the scheduled task survives restarts.
Metadata
Metadata
Assignees
Labels
No labels
Activity
juliusmarminge commentedon Nov 22, 2026
This is the same problem as #38123, so I'm closing it in favour of that one. If it is different, say how and it will be reopened.