Diagnosis: "build agent empty input" after orchestrator output

Status: Diagnosis only — no code change yet. Date: 2026-07-19 Related PR: #818 (fix/preset-tui-slash-command) — same root class as the original /preset fix. Suspected sibling bug reported by user: During superpowers / brainstorm skill conversations, when the orchestrator asks for confirmation or work is interrupted (subagent completes, background task finishes), a build agent turn sometimes appears with an empty user input.

TL;DR

The build agent turn with empty input is the same class of bug as the original /preset issue fixed in #818: a plugin hook calls sessionSdk.promptAsync({ body: { parts: [createInternalAgentTextPart(...)] } }) without specifying an agent field. opencode then resolves the agent via agents.defaultInfo(), which falls back to the built-in build agent whenever default_agent is unset, user-overridden, or not effectively applied. The synthetic: true flag hides the injected text from the TUI, so the user perceives the build turn as having "empty input."

Update (Issue #854): the incomplete-todo continuation path now passes agent: 'orchestrator', is an opt-in beta via backgroundJobs.continueOnIdle: true, and uses a process-local one-attempt gate. Remaining agent-less promptAsync call sites are interview/smartfetch (below).

Root cause (causal chain, cross-validated)

Orchestrator enters input-wait. After emitting a confirmation question (skill flow), the assistant turn finishes. opencode's per-session Runner transitions to Idle (packages/opencode/src/effect/runner.ts:115-138, packages/opencode/src/session/run-state.ts:60-63). The session is no longer "busy" from the Runner's perspective.
Plugin hook fires promptAsync with a synthetic part and historically no agent field. Remaining agent-less sites:
- src/interview/service.ts:622, 871, 933, 1007 — interview/skill flow injections (matches "brainstorm skill flow").
- Continuation nudge (continuation-evaluator.ts) previously matched "subagent completes" / "background task finishes"; it now sets agent: 'orchestrator' and only runs when continueOnIdle is enabled.
opencode does not guard promptAsync against busy/input-wait state. The HTTP handler at packages/opencode/src/server/routes/instance/httpapi/handlers/session.ts:311-329 does not call assertNotBusy and does not consult the Question service. It proceeds straight to promptSvc.prompt. The Runner, being Idle, immediately startRuns the new turn (runner.ts:131-134). No queue, no reject, no cancellation of the pending question.
Agent resolves to build. packages/opencode/src/session/prompt.ts:636-637:
```
const agentName = input.agent
const ag = agentName ? yield* agents.get(agentName) : yield* agents.defaultInfo()
```
When input.agent is omitted, opencode uses agents.defaultInfo() (packages/opencode/src/agent/agent.ts:328-340), which returns the first visible mode: "primary" agent — build (declared first in the agent registry, agent.ts:141-155). omos attempts to set default_agent = "orchestrator" via its config hook (src/index.ts:546-551), but only when default_agent is absent. build is selected whenever:
- config.setDefaultAgent === false (plugin config disables it)
- The user's opencode.json already sets a different default_agent
- The config hook didn't run or didn't apply (SDK/runtime version skew: plugin built against @opencode-ai/sdk v1.4.3, installed runtime v1.18.3)
- The orchestrator agent isn't registered at config-load time
"Empty input" is the synthetic flag's visual effect. synthetic: true only controls TUI visibility. packages/opencode/src/session/message-v2.ts:206 still includes synthetic text parts in model messages (the only filters are !part.ignored and part.text !== ""; there is no synthetic filter when building model messages). The user sees the build agent respond to a turn with no visible user message — perceived as "empty input." (createInternalAgentTextPart appends a \n marker, so the text is non-empty from the LLM's perspective.)
Session agent is durably corrupted. packages/opencode/src/session/prompt.ts:672-689 compares current.agent !== info.agent and, if they differ, calls sessions.setAgentModel({ agent: info.agent, ... }). A single agent-less promptAsync that resolves to build permanently rewrites the session's agent to build — every subsequent turn also routes to build until explicitly reset.

Confirmed-affected call sites

File:line	Trigger	Body omits `agent`?	Gate
`src/hooks/task-session-manager/continuation-evaluator.ts` (`promptAsync`)	`session.idle` / `session.status(idle)` on orchestrator session with incomplete todos when the opt-in beta `backgroundJobs.continueOnIdle` is `true`	No (`agent: 'orchestrator'`)	`continueOnIdle`, process-local one-attempt gate (reserve→commit), `hasInputWait`, `isCurrentContinuation`, `isFallbackInProgress`, `backgroundJobBoard.hasTerminalUnreconciled`, malformed/active SDK short-circuits
`src/interview/service.ts:622`	User submits interview dashboard input	Yes	`sessionBusy` lock, interview active state
`src/interview/service.ts:871`	User submits interview chat	Yes	same
`src/interview/service.ts:933`	User submits interview answer	Yes	same
`src/interview/service.ts:1007`	User submits interview comment	Yes	same
`src/interview/service.ts:504`	Interview URL notification	Yes (but `noReply: true`, non-synthetic text)	none
`src/tools/smartfetch/secondary-model.ts:252`	Smartfetch secondary model query	Yes	none

Correct pattern (for comparison)

src/hooks/foreground-fallback/index.ts:635-639 explicitly includes the agent:

const promptBody = {
  parts: lastUser.parts,
  model: ref,
  ...(agentName ? { agent: agentName } : {}),
};

This is the pattern every promptAsync caller in omos should follow.

Why the `hasInputWait` gate in task-session-manager is not sufficient

The gate exists and works in the common case (see continuation-evaluator.ts and task-session-manager/index.test.ts continuation cases). Notes:

Continuation is opt-in beta. backgroundJobs.continueOnIdle defaults to false; set it to true to enable continuation SDK calls. Idle reconciliation remains active either way. When enabled, a process-local reserve/commit gate allows at most one promptAsync per session epoch between real user messages.
Documented race window (when enabled). IDLE_RECONCILE_DELAY_MS = 2_000. The idle-reconciliation comment admits late completions can still race the window. If session.idle fires and the timer expires before question.asked is delivered, and the SDK liveness reads resolve before the wait is tracked, a nudge can still fire.
Input-wait is not the only trigger. The interview/skill path (src/interview/service.ts) does not consult hasInputWait at all — it injects on user dashboard actions, which can happen while the orchestrator is mid-question.
Continuation path sets agent: 'orchestrator'. The historical missing- agent misroute on the continuation nudge is fixed in continuation-evaluator.ts. Remaining agent-less promptAsync call sites are outside that path (interview/smartfetch below).

Why this is the same class as the #818 `/preset` fix

#818's original bug: /preset used createInternalAgentTextPart() to trigger an LLM turn that was invisible in the TUI (synthetic: true). The fix moved /preset to pure TUI dialogs (src/tui-preset.ts uses only api.ui.dialog / DialogSelect / DialogPrompt / DialogConfirm — no promptAsync).

This bug: other hooks still use the same createInternalAgentTextPart + promptAsync pattern, and additionally omit the agent field, so the invisible turn routes to build instead of the orchestrator. Same shape: a synthetic part starting an invisible turn. Different symptom: build agent instead of orchestrator.

Fix directions (not implemented — awaiting decision)

Minimal fix

Continuation already passes agent: 'orchestrator'. Add the same to remaining agent-less promptAsync bodies:

src/interview/service.ts (dashboard/chat/answer/comment injectors)

This ensures interview injections always route to the orchestrator regardless of opencode's default_agent resolution, eliminating the path to build.

Hardening (optional, larger scope)

Input-wait guard on the interview/skill path. Consult hasInputWait (or an equivalent signal) before injecting in src/interview/service.ts. Do not inject while the orchestrator is waiting for user input.
Post-injection agent assertion. After each promptAsync, assert current.agent was not changed out from under the orchestrator; if it was, restore it via setAgentModel.
Investigate the default_agent application reliability on opencode v1.18.x. The plugin was built against @opencode-ai/sdk v1.4.3; the installed runtime is v1.18.3. The config hook's default_agent = 'orchestrator' mutation may not be applied reliably under this skew. (Note: #799 tracks the package upgrade.)
Shrink or eliminate the IDLE_RECONCILE_DELAY_MS race for sessions that have a pending question.asked / permission.asked.

Evidence index

omos source

Continuation nudge (fixed agent + opt-in beta + one-attempt gate): src/hooks/task-session-manager/continuation-evaluator.ts, continuation-attempt-gate.ts, backgroundJobs.continueOnIdle in src/config/schema.ts
Missing agent field (skill flow): src/interview/service.ts:622, 871, 933, 1007
Correct pattern for comparison: src/hooks/foreground-fallback/index.ts:635-639
omos sets default_agent only when absent: src/index.ts:546-551
createInternalAgentTextPart produces synthetic: true: src/utils/internal-initiator.ts:9-21
CONTINUATION_NUDGE is non-empty: src/hooks/task-session-manager/continuation-evaluator.ts
hasInputWait / continuation gates: continuation-evaluator.ts, input-wait-tracker.ts
IDLE_RECONCILE_DELAY_MS race window: src/hooks/task-session-manager/index.ts (IDLE_RECONCILE_DELAY_MS)
disableDefaultAgents preserves build and plan: src/cli/config-io.ts:564-600

opencode source (`anomalyco/opencode` @ `dev`)

promptAsync HTTP handler (no busy/input-wait guard): packages/opencode/src/server/routes/instance/httpapi/handlers/session.ts:311-329
prompt HTTP handler (no guard): same file, :295-309
Internal prompt always starts a new turn: packages/opencode/src/session/prompt.ts:1052-1071
Agent selection: input.agent ?? defaultInfo(): packages/opencode/src/session/prompt.ts:636-637
defaultInfo() → default_agent or first visible primary: packages/opencode/src/agent/agent.ts:328-340
build agent definition (default primary, first in registry): packages/opencode/src/agent/agent.ts:141-155
Destructive setAgentModel overwrite on agent change: packages/opencode/src/session/prompt.ts:672-689
Synthetic parts included in model messages: packages/opencode/src/session/message-v2.ts:206
Runner ensureRunning (Idle = run now, no queue): packages/opencode/src/effect/runner.ts:115-138
Runner Idle transition on turn end: packages/opencode/src/session/run-state.ts:60-63
assertNotBusy (NOT used by prompt/promptAsync): packages/opencode/src/session/run-state.ts:71-75

SDK types

default_agent doc: "Falls back to 'build' if not set or invalid": node_modules/@opencode-ai/sdk/dist/v2/gen/types.gen.d.ts:1255-1257
SessionPromptAsyncData.body.agent? is optional: node_modules/@opencode-ai/sdk/dist/v2/gen/types.gen.d.ts:3241-3269
build is a built-in agent: node_modules/@opencode-ai/sdk/dist/v2/gen/types.gen.d.ts:1273-1279

Open questions for the fix

Should the fix be a new PR, or amended into #818? (#818 is currently MERGEABLE / CLEAN / CI green; amending widens its scope and may delay merge.)
Is the interview/skill path expected to always route to the orchestrator, or could it intentionally target a different agent in some flows?
Should we also harden the default_agent application (fix direction #3) as part of this work, or track it separately under #799?

This report is diagnostic only. No code was changed. The fix awaits the user's decision on scope and PR strategy.

build-agent-empty-input-diagnosis.md 12 KB Kalıcı Bağlantı Geçmiş Ham