sorokin.oleg.edu/learnharness
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases 6 Wiki Activity

fix: reasoning always renders above its message; dock drops delivered subagents at delivery

Browse Source 
TWO owner-reported bugs from the live red-circle session
(chat_muhpyoll_27305e23), both investigated to root cause first:

1. THINKING BLOCKS UNDER THE MESSAGE. GLM emits trailing reasoning
   tokens, so the persisted content lands [text, thinking] (4 of the 6
   thinking-bearing messages from [21] on in that session) - and the
   view canon presented arrival order verbatim, rendering the reasoning
   UNDER its own message. Fix: projectView stable-partitions each
   assistant message - reasoning parts first (their own order kept),
   then text/tool parts. Owner ruling 2026-09-26: the reasoning belongs
   above, always. Covers live partials and settled records alike (one
   projection canon). NOT addressed (owner ruling): the plan_tools
   "missing" claim in the same session - wire-verified 11 tools incl.
   plan_init; GLM confabulation, no mitigation wanted.

2. DOCK CARDS LINGERING AFTER DELIVERY. /api/activity lists jobs with
   status "running", and the status flipped only in settleSteers at the
   DELIVERING TURN'S onFinish - so a subagent whose card was already
   visible in the transcript (marker injected mid-run) stayed in the
   background dock until the teacher finished teaching. Fix: flip the
   job at the moment the result reaches the learner - noteSteerVisible
   (jobs.ts) called from BOTH visibility seams: the marker's message_end
   wire event (live steer injection, ui-stream) and the write-ahead
   persist (the wake's input appends, chat-handler). finalizeJob is
   idempotent; settleSteers/runner-200 re-run it as a no-op, and the
   registry's delivery promise still settles exactly as before.

Tests: 1001/1001 - view-canon hoist pin (trailing [text, think, call]
projects reasoning-first; think-first unchanged), noteSteerVisible unit
pin (flip at the seam, idempotent, handoff no-op), and the mid-run
steering e2e now holds the turn open past the injection and pins the
job flipping done at the marker's message_end WHILE the turn still
streams. Live-verified read-only on the owner's server: the original
session reloads with every reasoning trigger above its message.
This commit is contained in:
amosblomqvist
2026-09-26 05:46:33 +03:00
parent ad0a6e5201
commit b84ab1cdc0
6 changed files with 129 additions and 8 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/server/pi/ui-stream.ts
+7
View File
@@ -1,4 +1,6 @@
import type { Agent, AgentEvent, AgentMessage } from "@earendil-works/pi-agent-core";
import type { SteerMessage } from "../../core/contracts.ts";
import { noteSteerVisible } from "../jobs.ts";
/**
* THE NATIVE WIRE (Wave D, owner-approved 2026-09-26): one agent turn, streamed
@@ -123,6 +125,11 @@ export function runPiTurn(spec: PiTurnSpec): PiTurnHandle {
if (event.type === "message_end") {
const m = event.message as { role?: string; stopReason?: string } | undefined;
if (m?.role === "assistant" && m.stopReason === "aborted") sawAbort = true;
// Dock fix (owner ruling 2026-09-26): an injected steer marker's
// message_end = the card is visible to the learner — flip the
// job's bookkeeping now (the turn may keep teaching for minutes;
// settleSteers re-runs the flip as a no-op idempotently).
if (m?.role === "custom") noteSteerVisible(event.message as SteerMessage);
}
if (event.type === "agent_end") {
closeAll();