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

Wave B: the native message record — input events, projection, adapter death

Browse Source 
The chat record is now pi-native agent history (LhMessage[]), the AI SDK's
UIMessage transcript contract is gone from the server, and every client turn
arrives as ONE explicit input event.

Core (new):
- src/core/native-messages.ts — the LhMessage taxonomy (user/assistant/
  toolResult + lhMarker customs: verdict, probe-handoff, subagent-steer,
  subagent-abort, plan-block), constructors, and toLlmMessages — the single
  model-facing seam (convertToLlm): strips ids, flattens markers to user
  text, drops UI-only plan-blocks, synthesizes INTERRUPTED results for
  dangling aborted calls, never throws.
- src/core/projection.ts — projectNative: the one-way native → UIMessage
  projection (consecutive assistant+toolResult runs group into one
  assistant view message; toolResults attach backward by toolCallId;
  markers project to data parts; plan-blocks splice after the last
  successful plan_init). Every existing consumer reads through it.
- src/core/input-events.ts — the ChatInput union (text/answer/verdict/
  continue), parseChatInput validation, and inputAppends (answer targets
  the dangling question call or 400s; verdict = optional note user
  message + native verdict marker).

Server:
- chat-handler.ts — the turn root takes {chatId, input}; write-ahead
  persists input appends + claimed steers before the turn; claimSteers
  coalesces the whole queue with the handler's steerDelivered filter as
  the dedupe seam; teacher-vs-prober branch on hasProbeDone.
- turn-service.ts — persistTurn stamps ids, drops clean question
  placeholder toolResults (the record ends at the unanswered call),
  emits plan-block markers idempotently (plan-blocks.ts rewrite), and
  persists abort partials.
- ui-stream.ts — PiTurnSpec takes the native base; the wire dialect is
  unchanged. jobs.ts steers are native markers (steer_<jobId> dedup ids);
  waitForIdle reads danglingQuestionOf. teacher-agent.ts wires
  convertToLlm: toLlmMessages. record-adapter.ts DELETED wholesale.
- GET /api/chats/[id] serves messages: projectNative(record.messages).

Client:
- chat-transport.ts — the 409 retry is IDENTITY (re-POST the same event
  body after the turn settles — the event was never persisted
  pre-acceptance; the merge logic is gone). The transport consumes the
  stashed input event via prepareSendMessagesRequest and throws loudly on
  a missing stash.
- chat.tsx — every send call site stashes exactly one ChatInput event
  (text, answer, verdict, continue) before invoking the SDK send.

Tests: full native migration (chat-flow rewritten to input events and
native record pins; jobs/extraction/demos/graph/usage/plan-blocks/
probe-done/reattach-transport converted; record-adapter.test.ts deleted;
chat-transport tests pin the identity retry). 982/982, typecheck, lint,
build all green.
This commit is contained in:
amosblomqvist
2026-09-26 00:37:46 +03:00
parent b87a6925af
commit 7ab47d4fcb
36 changed files with 2410 additions and 1447 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/agent/subagents/extraction.ts
+10 -6
View File
@@ -28,6 +28,7 @@ import { getGraph, mergeFindings, type EmissionStats, type GraphDeps } from "../
import type { GraphStore } from "../../core/ports.ts";
import { settleGraphLayout } from "../../core/layout-settle.ts";
import { batchRuns, batchTag, coverageOf, messageViewLines } from "../../core/coverage.ts";
import { projectNative } from "../../core/projection.ts";
/**
* The extraction pipeline (vision §1 construction, 2026-09-02 redesign) — the
@@ -92,10 +93,12 @@ import { batchRuns, batchTag, coverageOf, messageViewLines } from "../../core/co
export function transcriptView(record: ChatRecord): string {
// The shared seam (core/coverage.ts messageViewLines): the LLM's prompt
// view and the coverage fingerprint use ONE line builder, so "covered"
// means — exactly — what the extractor saw.
// means — exactly — what the extractor saw. Wave B: the record is native
// history — the extraction view reads through the one-way projection.
const lines = [`SESSION: ${record.title}`];
record.messages.forEach((message, index) => {
lines.push(...messageViewLines(message as UIMessage, (message as UIMessage).id || `m${index}`));
const projected = projectNative(record.messages);
projected.forEach((message, index) => {
lines.push(...messageViewLines(message, message.id || `m${index}`));
});
return lines.join("\n");
}
@@ -868,7 +871,8 @@ export function createExtractionPipeline(deps: ExtractionPipelineDeps): Extracti
// engines AND their providers' keys; a refusal lands BEFORE any
// stage runs (typed — the trigger skips, routes surface 503).
refuseIfStagesUnready();
const covered = coverageOf(record.messages as UIMessage[], graphStore.readCoverage(), chatId);
const projected = projectNative(record.messages);
const covered = coverageOf(projected, graphStore.readCoverage(), chatId);
if (covered.uncovered.length === 0) {
console.info(`[extraction ${chatId}] nothing uncovered — no-op`);
return { nodes: 0, edges: 0, events: 0 };
@@ -883,11 +887,11 @@ export function createExtractionPipeline(deps: ExtractionPipelineDeps): Extracti
// The batch's prompt view: SESSION header + the batch's lines,
// same provenance [id] tags the fingerprints used (coverage
// and prompt share ONE line builder — core/coverage.ts).
const mids = record.messages.map((m, i) => ((m as UIMessage).id || `m${i}`) as string);
const mids = projected.map((m, i) => (m.id || `m${i}`) as string);
const view = [
`SESSION: ${record.title}`,
...batch.flatMap((slot) => {
const idx = record.messages.indexOf(slot.message);
const idx = projected.indexOf(slot.message);
return messageViewLines(slot.message, mids[idx] ?? slot.message.id);
}),
].join("\n");
src/agent/subagents/runner.ts
+10 -4
View File
@@ -1,4 +1,4 @@
import { isToolUIPart, type UIMessage } from "ai";
import { isToolUIPart } from "ai";
import type { PiEngine } from "../models.ts";
import type { SearchProvider } from "../../core/search.ts";
import {
@@ -14,7 +14,7 @@ import {
type PdfToolInput,
type ResearchResult,
type ResearchToolInput,
type SubagentSteerMessage,
type SteerMessage,
type UsageStepLike,
type VisualResult,
type VisualToolInput,
@@ -23,6 +23,8 @@ import { registerJobAbort, unregisterJobAbort } from "./aborts.ts";
import type { ModelRole } from "../../../config.ts";
import { ROLE_SLOT, modelsNotConfigured, type ModelSlotName, type ProviderKeyMissing } from "../../core/settings.ts";
import type { ChatStore, JobStore } from "../../core/ports.ts";
import { projectNative } from "../../core/projection.ts";
import type { UIMessage } from "ai";
import { runDemoBuilder } from "./demo-builder.ts";
import { runPdfReaderAgent } from "./pdf-reader.ts";
import { runResearcherAgent } from "./researcher.ts";
@@ -69,7 +71,7 @@ export interface JobRunnerDeps {
* steer is delivered (the wake turn completed and the job status
* flipped); rejects when delivery fails unrecoverably.
*/
deliver: (steer: SubagentSteerMessage) => Promise<void>;
deliver: (steer: SteerMessage) => Promise<void>;
/** Settings-resolved engine per role (model slots wave, 2026-09-20):
* the role-override → slot chain, read FRESH per job (the server
* wiring passes effectiveSettings-based resolution). Test overrides
@@ -274,7 +276,11 @@ export function createJobRunner(deps: JobRunnerDeps): JobRunner {
if (input) return input;
const chat = chatStore.getChat(job.chatId);
if (!chat) return undefined;
return (findToolPart(chat.messages, job.toolCallId)?.input ?? undefined) as JobToolInput | undefined;
// Wave B: the record is native history — the transcript scan reads
// through the projection (same shape the old record held).
return (findToolPart(projectNative(chat.messages), job.toolCallId)?.input ?? undefined) as
| JobToolInput
| undefined;
}
// ─── shared plumbing ────────────────────────────────────────────────────
src/agent/teacher-agent.ts
+9 -1
View File
@@ -15,6 +15,7 @@ import {
} from "./tools.ts";
import { quizInputObject, quizInputProblems } from "../core/question-core.ts";
import { QUESTION_TOOLS, type EnabledTools } from "../core/pending.ts";
import { toLlmMessages } from "../core/native-messages.ts";
/**
* The two teaching-side agents (docs/probe-teach-split.md): the PROBER and
@@ -89,7 +90,7 @@ export type StepUsageCollector = (usage: Usage) => void;
export interface AgentBuildOptions {
/** Model + stream function — tests inject a scripted faux engine. */
engine?: PiEngine;
/** The initial transcript, as pi messages (recordToPiMessages). */
/** The initial transcript, as pi messages (the record IS agent state). */
messages?: AgentMessage[];
/** Money channel: one call per model request. */
onUsage?: StepUsageCollector;
@@ -164,6 +165,13 @@ function baseAgentOptions(kind: AgentKind, opts: AgentBuildOptions, tools: Agent
},
streamFn: engine.streamFn,
shouldStopAfterTurn: stopper,
// Wave B (owner-approved 2026-09-26): the model-facing conversion is
// THE app seam — markers flatten to their canonical user text
// (verdictText / HANDOFF_TEXT / steer text, all single-sourced),
// UI-only markers (plan-block) drop, the app's id fields strip, and
// an abort's dangling question call gets its INTERRUPTED result. It
// must never throw (pi's convertToLlm contract).
convertToLlm: toLlmMessages,
});
if (opts.onUsage) {
agent.subscribe((event: AgentEvent) => {
src/agent/tools/probe.ts
+9 -11
View File
@@ -1,10 +1,9 @@
import type { AgentTool } from "@earendil-works/pi-agent-core";
import { serverExecute } from "../tool-execute.ts";
import { probeDoneInputSchema, type ProbeDoneResult } from "../../core/contracts.ts";
import type { UIMessage } from "ai";
import { findPendingToolCall } from "../../core/pending.ts";
import { danglingQuestionOf, hasProbeDone, type LhMessage } from "../../core/native-messages.ts";
import type { ChatStore } from "../../core/ports.ts";
import { buildProbeHandoffSteer, type ProbeHandoffSteerMessage } from "../../core/contracts.ts";
import { buildProbeHandoffSteer, type SteerMessage } from "../../core/contracts.ts";
/** What the probe_done tool closes over. */
export interface ProbeDoneContext {
@@ -13,15 +12,14 @@ export interface ProbeDoneContext {
chatStore: ChatStore;
/** Queue the handoff steer (the wake pipeline — src/server/jobs.ts).
* Injected so this module never imports the server seam. */
deliver: (steer: ProbeHandoffSteerMessage) => void;
deliver: (steer: SteerMessage) => void;
}
/** Does the transcript already carry a concluded probe (any state)? */
function probeConcluded(messages: UIMessage[]): boolean {
for (const m of messages) {
if (m.parts.some((p) => p.type === "tool-probe_done")) return true;
}
return false;
/** Does the transcript already carry a concluded probe (any state)?
* Wave B: the native scan (any probe_done tool call, any state) — same
* semantics the UIMessage part scan had. */
function probeConcluded(messages: readonly LhMessage[]): boolean {
return hasProbeDone(messages);
}
/**
@@ -64,7 +62,7 @@ export function probeDoneTool(ctx: ProbeDoneContext): Record<string, AgentTool<a
const record = ctx.chatStore.getChat(ctx.chatId);
if (!record) return { ok: false, error: `Chat not found: ${ctx.chatId}` };
if (findPendingToolCall(record.messages) !== null) {
if (danglingQuestionOf(record.messages) !== null) {
return {
ok: false,
error: