fix(agent-loop): quiesce scheduler failures

.agents/notes/implemented/feature/2026-07-10-parallel-tool-call-execution.md

@@ -10,7 +10,7 @@ An assistant message may contain several sibling `tool-call` blocks. Running the
 
 Concurrency is a host scheduling concern, not model-facing tool metadata. The loop needs to decide which calls may overlap without hardcoding tool names or exposing scheduler policy in the JSON schema.
 
-The session log remains authoritative: every started call has an audit event, every started call receives a result, and model history observes results in the original call order regardless of completion order.
+The session log remains authoritative: every started call has an audit event, ordinary completion and cancellation pair calls with results, and model history observes committed results in the original call order regardless of completion order.
 
 ## Decision
 
@@ -46,7 +46,7 @@ Only dispatch and the tool body overlap. `tools/pre-execute` and `tools/post-exe
 
 Each started call appends `tool/call` immediately before its pre-execute gate. Completed dispatches occupy model-order slots, and a commit cursor appends `tool/result` and collects `additionalContexts` only when the next slot is ready. Live surfaces may show several pending calls, but results and post-tool context remain model-ordered.
 
-An abort before a group starts records no calls from that group. An abort during a group stops replenishment, waits for already-started calls, commits their results in order, drains accepted batch context after those results, and then ends the step through the existing abort path. Calls that never start have no audit event.
+An abort before a group starts records no calls from that group. An abort during a group stops replenishment, waits for already-started calls, commits their results in order, drains accepted batch context after those results, and then ends the step through the existing abort path. Calls that never start have no audit event. An unexpected scheduler failure stops new dispatches, waits for every already-started dispatch to settle, and rethrows the first failure. Because that failure is terminal internal state rather than a tool outcome, the loop does not invent tool results for rejected or uncommitted calls.
 
 Code Mode remains outside this scheduler because the model emits one native `run_code` call. `run_code` and its internal dispatch queue remain serial; native sibling calls in `mode: 'both'` use the normal scheduler.
 
@@ -66,7 +66,7 @@ Filesystem read relies on a narrow recorder exception: its synchronous observati
 
 ## Verification
 
-Unit coverage pins fail-closed classification, typed argument validation, grouping, barriers, live reclassification after registry replacement, the rolling cap, distinct execution objects, middleware order, ordered results and context, and abort draining. First-party tests pin each parallel declaration.
+Unit coverage pins fail-closed classification, typed argument validation, grouping, barriers, live reclassification after registry replacement, the rolling cap, distinct execution objects, middleware order, ordered results and context, abort draining, and scheduler-failure quiescence. First-party tests pin each parallel declaration.
 
 Snapshot coverage pins the visible multi-call transcript: pending calls may overlap while completed results remain model-ordered. Code Mode coverage pins its serial boundary. No provider-backed e2e is required because scheduling is deterministic loop behavior.
 
@@ -84,6 +84,8 @@ Snapshot coverage pins the visible multi-call transcript: pending calls may over
 
 **Expose staged methods or a scheduling waterfall.** Public `prepare` / `dispatch` / `finalize` methods or a `tools/execution-mode` event add extension surface before another consumer needs it. The loop uses an internal scheduler view, while `executionMode(exec)` leaves an insertion point for a policy seam.
 
+**Convert scheduler failures into tool results.** AgentLoop cannot determine whether a rejected dispatch invoked the tool body; ToolRegistry owns body-invocation state and typed tool outcomes. Internal scheduler failures therefore remain terminal instead of being reclassified as `ABORTED` results.
+
 **Start calls while the model streams.** This may reduce latency further but changes assistant-message authority, replay, and call/result pairing. The scheduler starts only after the assistant message is complete.
 
 **Use fixed-size windows.** Waiting for every call in one window before starting the next leaves capacity idle behind a slow call. The rolling pool preserves the cap without that delay.
@@ -101,3 +103,5 @@ Ordered commits may hold a fast result behind a slow earlier sibling. This prese
 Concurrent external calls can compete for quota or process capacity. Providers own their capacity controls; the loop cap only limits calls from one agent step.
 
 Tool registration is a scheduling boundary. Registry mutations affect not-yet-started calls because the scheduler reclassifies after each barrier and before every pool replenishment. Already-started calls retain the scheduling decision under which they entered the pool.
+
+A terminal scheduler failure may leave recorded calls without results before the failed step closes. Waiting for live dispatches preserves quiescence without misreporting those internal failures as tool outcomes.