What happened, and what did you expect?
Several background fixer tasks entered running, then stopped without:
a terminal result,
a tool invocation,
an attributable child process,
a source/fixture hash change, or
a checkpoint/blocker report.
I expected each task to either execute and return a terminal result, or return a structured startup/runtime error.
background task lifecycle/result delivery appears unreliable. The same Windows checkout successfully runs direct commands, earlier Fixer tasks, the compiler, verifier, updater, and clangd. The failure occurs before the affected worker performs its first observable action.
Expected: task status and terminal result accurately reflect execution, with startup/provider errors exposed.
Actual: task stops silently; no work occurs.
Steps to reproduce
Steps to reproduce
Open a Windows workspace with a large C++ source file (~5.5 MB) and paired fixture.
Dispatch a background fixer task to make a bounded source/fixture edit.
Observe that some tasks stop without a terminal payload or filesystem/process evidence.
Repeat with narrowed scopes; failures remain intermittent.
oh-my-opencode.json
OpenCode version
1.18.13
oh-my-opencode-slim version
2.2.14
Operating system
Windows 11
Logs, screenshots, or extra context
What happened, and what did you expect?
Several background fixer tasks entered running, then stopped without:
a terminal result,
a tool invocation,
an attributable child process,
a source/fixture hash change, or
a checkpoint/blocker report.
I expected each task to either execute and return a terminal result, or return a structured startup/runtime error.
background task lifecycle/result delivery appears unreliable. The same Windows checkout successfully runs direct commands, earlier Fixer tasks, the compiler, verifier, updater, and clangd. The failure occurs before the affected worker performs its first observable action.
Expected: task status and terminal result accurately reflect execution, with startup/provider errors exposed.
Actual: task stops silently; no work occurs.
Steps to reproduce
Steps to reproduce
Open a Windows workspace with a large C++ source file (~5.5 MB) and paired fixture.
Dispatch a background fixer task to make a bounded source/fixture edit.
Observe that some tasks stop without a terminal payload or filesystem/process evidence.
Repeat with narrowed scopes; failures remain intermittent.
oh-my-opencode.json
{ "$schema": "https://unpkg.com/oh-my-opencode-slim@latest/oh-my-opencode-slim.schema.json", "preset": "omniroute-hybrid", "fallback": { "enabled": true, "timeoutMs": 8000, "retryDelayMs": 250, "maxRetries": 2, "retry_on_empty": true }, "backgroundJobs": { "maxSessionsPerAgent": 6 }, "presets": { "openai": { "orchestrator": { "model": "openai/gpt-5.6-terra", "variant": "xhigh", "skills": [ "*" ], "mcps": ["gh_grep"] }, "oracle": { "model": "openai/gpt-5.6-sol", "variant": "xhigh", "skills": [ "simplify" ], "mcps": [] }, "librarian": { "model": "openai/gpt-5.6-luna", "variant": "low", "skills": [], "mcps": [ "context7", "gh_grep" ] }, "explorer": { "model": "openai/gpt-5.6-luna", "variant": "low", "skills": [], "mcps": [] }, "designer": { "model": "openai/gpt-5.6-luna", "variant": "medium", "skills": [], "mcps": [] }, "fixer": { "model": "openai/gpt-5.6-luna", "variant": "xhigh", "skills": [], "mcps": [] } }, "opencode-go": { "orchestrator": { "model": "opencode-go/minimax-m3", "variant": "thinking", "skills": [ "*" ], "mcps": ["gh_grep"] }, "oracle": { "model": "opencode-go/qwen3.7-max", "variant": "max", "skills": [ "simplify" ], "mcps": [] }, "explorer": { "model": "opencode-go/deepseek-v4-flash", "variant": "high", "skills": [], "mcps": [] }, "librarian": { "model": "opencode-go/deepseek-v4-flash", "variant": "high", "skills": [], "mcps": [ "context7", "gh_grep" ] }, "designer": { "model": "opencode-go/kimi-k2.7-code", "skills": [], "mcps": [] }, "fixer": { "model": "opencode-go/deepseek-v4-flash", "variant": "high", "skills": [], "mcps": [] }, "observer": { "model": "opencode-go/mimo-v2.5", "skills": [], "mcps": [] } }, "omniroute-hybrid": { "orchestrator": { "model": [ "openai/gpt-5.6-terra", "opencode-go/deepseek-v4-pro", "omniroute/Orchestrator" ], "variant": "high", "skills": ["no-ai-slop"], "mcps": [] }, "oracle": { "model": [ "openai/gpt-5.6-sol", "opencode-go/deepseek-v4-pro", "omniroute/Oracle" ], "variant": "max", "skills": ["simplify"], "mcps": [] }, "explorer": { "model": [ "opencode-go/deepseek-v4-flash", "openai/gpt-5.6-luna", "omniroute/Explorer" ], "variant": "high", "skills": [], "mcps": [] }, "librarian": { "model": [ "opencode-go/deepseek-v4-flash", "openai/gpt-5.6-luna", "omniroute/Librarian" ], "variant": "high", "skills": [], "mcps": ["context7", "gh_grep"] }, "designer": { "model": [ "openai/gpt-5.6-luna", "opencode-go/kimi-k2.7-code", "omniroute/Designer" ], "variant": "medium", "skills": [], "mcps": [] }, "fixer": { "model": [ "opencode-go/deepseek-v4-flash", "openai/gpt-5.6-luna", "omniroute/Fixer" ], "variant": "high", "skills": [], "mcps": [] }, "observer": { "model": [ "openai/gpt-5.6-luna", "opencode-go/mimo-v2.5", "omniroute/Observer" ], "skills": [], "mcps": [] }, "council": { "model": "openai/gpt-5.6-sol", "variant": "high", "skills": ["simplify"], "mcps": [] } } }, "council": { "default_preset": "omniroute-hybrid", "presets": { "omniroute-hybrid": { "architect": { "model": "openai/gpt-5.6-terra", "variant": "medium", "prompt": "Develop the strongest system-level design and identify key trade-offs." }, "challenger": { "model": "opencode-go/deepseek-v4-pro", "variant": "max", "prompt": "Challenge the proposed approach. Find failure modes, missing constraints, and better alternatives." }, "skeptic": { "model": "omniroute/Council", "variant": "max", "prompt": "Act as an independent skeptic. Identify assumptions, edge cases, and practical risks." } } } }, "disabled_agents": [] }OpenCode version
1.18.13
oh-my-opencode-slim version
2.2.14
Operating system
Windows 11
Logs, screenshots, or extra context