drive: cloud run 8fcffb06

ops/NEXT.md

@@ -1,87 +1,79 @@
 # NEXT — work package for this tick
 
-**Scope:** Build a minimal agent worker in the SDK. CODE task, SDK-side.
+**Gate:** Gate 2 (proactive agent workload — hn-monitor runs as a relayflow)
 
-This run is pinned to **gate 3** and must not work on any other gate.
+**Scope:** Build sub-PR A of the Gate 2 push: a real `hn-monitor` polling runner in the SDK. CODE task, `sdk/src/`-side. This is a scaffolding PR — proof that the workload EXECUTES end-to-end is deliberately deferred to sub-PR B (integration test).
 
 ## Objective
 
-Promote the throwaway worker the tests already build into a real SDK component
-that can execute agent steps by running their declared CLI as a subprocess.
+Add `sdk/src/hn-monitor-runner.ts` that composes existing pieces (JournalClient from `sdk/src/journal-client.ts`, AgentWorker from `sdk/src/worker.ts`, pollHackerNewsOnce from `sdk/src/hn-poller.ts`) into a continuous runner for the hn-monitor workload.
 
-## Context
+This addresses five findings from closed PR #83:
 
-Nothing in this repo can execute an agent step. Searching for `workerAttach` /
-`step.complete` finds only TESTS (`sdk/tests/live-kernel.test.ts`,
-`journal-client.test.ts`, `journal-client-loopback.ts`) and the protocol
-definitions. `sdk/src/cli/run.ts` only OBSERVES worker leases and waits for one
-that never arrives.
+1. **Fail-closed on journal errors** — only fetch errors may be swallowed; journal write failures MUST throw
+2. **AgentWorker.close() must release the worker** — either add workerRelease verb or document what close() does NOT do
+3. **Class field declaration order** — all fields before constructor
+4. **Signal handlers opt-in via AbortSignal** — no process-wide signal handlers
+5. **Test coverage for pollError branch** — loop survives fetch throw AND terminates on journal throw
 
-The kernel's dispatch, lease and claim machinery is real and tested. The worker
-side of the protocol is simply unimplemented, and that is what blocks gate 2
-("a workload RUNS as a relayflow" — today a run can only be shown CREATED) and
-gate 3 ("every claim/lease/retry served by the kernel").
+## Context
 
-`sdk/tests/live-kernel.test.ts` around the `live-manual-agent` case (line 288)
-shows the whole shape: connect, `hello`, `workerAttach` with pins, receive
-`step.dispatch`, act, complete. The protocol is already proven there.
+RFC-0001 §3 gate 2 is done when "hn-monitor runs as a relayflow in production, triggered by its real events, with zero bespoke persistence." Every primitive already exists: event triggers (PR #14), the flow spec (`testdata/hn-monitor.flow.yaml`), the poller (`sdk/src/hn-poller.ts`), the agent worker (`sdk/src/worker.ts` from PR #53), and a one-shot demo (`sdk/src/demo-hn-monitor.ts`). Nothing has run them together as a continuous workload. This PR fixes that.
 
 ## Files in scope
 
-- `sdk/src/worker.ts` — new file, the worker implementation
-- `sdk/src/index.ts` — export the worker
-- `sdk/tests/live-kernel.test.ts` OR a new test file — add a test that runs a
-  real flow with an agent step end to end against a live `relayflowd`, with
-  this worker attached, and asserts the step reaches `done`.
+- `sdk/src/hn-monitor-runner.ts` — NEW, the continuous runner
+- `sdk/src/worker.ts` — MAY modify close() per finding #2 (add workerRelease or document)
+- `sdk/src/protocol.ts` — IF workerRelease verb needs to be added
+- `sdk/tests/hn-monitor-runner.test.ts` — NEW, test coverage
+- `sdk/src/index.ts` — export HnMonitorRunner
 
 ## Definition of done
 
 ALL of the following must hold:
 
-1. The worker in `sdk/src/worker.ts`, exported from `sdk/src/index.ts`
-
-2. A test that runs a real flow with an agent step end to end against a live
-   `relayflowd`, with this worker attached, and asserts the step reaches
-   `done`. `sdk/tests/live-kernel.test.ts` already starts a daemon — follow
-   that pattern.
-
-3. **The worker must attach BEFORE the run starts.** A run that finds no worker
-   parks, and attaching afterwards does not re-drive it — `run.resume` is what
-   picks a parked run back up. That contract is pinned in the live-kernel
-   suite; do not fight it.
-
-4. The worker must:
-   - attach for `agent` steps with the pins it holds
-   - on `step.dispatch`, run the step's declared `cli` as a subprocess
-   - report the result back through the existing protocol (`step.complete`, and
-     the failure path when the CLI exits nonzero)
-   - nothing speculative: no retries of its own, no scheduling, no LLM calls.
-     The kernel owns retry and lease policy — do not reimplement it.
-
-5. `cd sdk && npm test` must be green. Run it and paste the literal command and
-   output tail showing test counts.
-
-6. `cd kernel && sh ../ops/cargo.sh test` must be green. Run it and paste the
-   literal command and output tail showing test counts.
-
-7. EVERY new test confirmed to FAIL against current code, with the literal
-   failing output quoted in the summary.
-
-8. As your LAST action, run `git status --porcelain` and paste it.
+1. `sdk/src/hn-monitor-runner.ts` exists and exports `HnMonitorRunner`
+2. Exported from `sdk/src/index.ts`
+3. Runner constructs JournalClient and AgentWorker
+4. **Worker attach happens BEFORE first poll** (a parked run needs run.resume)
+5. Loop: pollHackerNewsOnce → sleep POLL_INTERVAL_MS (env-configurable, default 60000) → repeat
+6. **Fetch errors caught and handled** — loop continues to next tick
+7. **Journal errors MUST throw and terminate the runner** (fail-closed)
+8. Clean shutdown on AbortSignal.abort (drain in-flight, close client, release/document worker)
+9. `sdk/src/worker.ts` — either close() calls workerRelease, OR one-line comment naming what close() does NOT do
+10. `sdk/src/protocol.ts` — IF workerRelease added, include request/response definitions
+11. `sdk/tests/hn-monitor-runner.test.ts` covers ALL of:
+    - fake fetch + mock journal → runner submits event on each tick
+    - abort signal triggers clean shutdown within one tick (worker released/documented)
+    - worker attach before first poll
+    - **fetch throw → loop survives** (onPollError called, next tick runs)
+    - **journal throw → loop TERMINATES** (runner.run() rejects)
+12. EVERY new test confirmed to FAIL against current code — comment out source, capture failure output
+13. PR body explicitly names non-goals: integration test (sub-PR B), CLI wrapper (sub-PR C), gate-2 declaration (sub-PR D)
+
+Commands that MUST pass with literal output captured:
+
+```bash
+cd sdk && npm test
+```
+
+Exit 0, all tests passing.
+
+```bash
+git status --porcelain
+```
+
+Shows only intended modifications. Run this as LAST action.
 
 ## Explicitly OUT of scope
 
-- LLM steps — not in the gate 3 scope
-- Retry logic in the worker — the kernel owns retry policy
-- Scheduling or lease management — the kernel owns lease policy
-- Optimizations, abstractions, or speculative features
-- Changes to the kernel
-- Changes to existing tests (except adding new test cases)
-- Work on any gate other than gate 3
-
-## If blocked
-
-If gate 3 is genuinely unreachable from the current state, write
-ops/NEEDS_HUMAN.md saying exactly why and still end with ASSESS_DONE. Do not
-silently substitute different work: a run that reports progress on the wrong
-gate is worse than one that reports it is blocked.
+- Proving the workload actually executes end-to-end (sub-PR B: integration test with real relayflowd)
+- CLI wrapper `flows hn-monitor start` (sub-PR C)
+- ops/STATE.md gate-2 GREEN declaration (sub-PR D)
+- `.github/workflows/*` — no GHA changes
+- `kernel/*` — kernel side already works via PR #14
+- `workflows/*.yaml` — for later sub-PRs
+- `ops/AUTODRIVE_BRIEF.md` — chief owns this file
+- Scheduling logic beyond the sleep — kernel owns retry/dedupe policy
+- LLM calls — runner is glue, not a reviewer
+- Rewriting worker.ts attach/dispatch/complete flow (PR #53 closed this)

sdk/src/hn-monitor-runner.ts

@@ -0,0 +1,105 @@
+import { JournalClient } from './journal-client.js';
+import { pollHackerNewsOnce, type Fetcher } from './hn-poller.js';
+import type { Pins } from './protocol.js';
+import { AgentWorker } from './worker.js';
+
+const DEFAULT_POLL_INTERVAL_MS = 60_000;
+
+const defaultFetcher: Fetcher = async (url) => {
+  const response = await fetch(url);
+  if (!response.ok) throw new Error(`HN fetch failed: HTTP ${response.status}`);
+  return response.text();
+};
+
+export interface HnMonitorRunnerOptions {
+  socketPath: string;
+  spec: unknown;
+  workerId: string;
+  pins: Pins;
+  signal?: AbortSignal;
+  pollIntervalMs?: number;
+  fetcher?: Fetcher;
+  storyLimit?: number;
+  onPollError?: (error: unknown) => void;
+}
+
+class FetchError {
+  constructor(readonly cause: unknown) {}
+}
+
+/** Continuously submits Hacker News events and serves their agent steps. */
+export class HnMonitorRunner {
+  private readonly client: JournalClient;
+  private readonly worker: AgentWorker;
+  private readonly pollIntervalMs: number;
+
+  constructor(private readonly options: HnMonitorRunnerOptions) {
+    this.client = new JournalClient(options.socketPath);
+    this.worker = new AgentWorker(this.client, {
+      workerId: options.workerId,
+      pins: options.pins,
+    });
+    this.pollIntervalMs = options.pollIntervalMs ?? pollIntervalFromEnv();
+  }
+
+  async run(): Promise<void> {
+    try {
+      await this.client.connect();
+      await this.worker.attach();
+      while (!this.options.signal?.aborted) {
+        await this.pollOnce();
+        await delay(this.pollIntervalMs, this.options.signal);
+      }
+    } finally {
+      this.worker.close();
+      this.client.close();
+    }
+  }
+
+  private async pollOnce(): Promise<void> {
+    const fetcher = wrapFetcher(this.options.fetcher ?? defaultFetcher);
+    try {
+      await pollHackerNewsOnce(this.options.spec, this.client, {
+        fetcher,
+        storyLimit: this.options.storyLimit,
+      });
+    } catch (error) {
+      if (!(error instanceof FetchError)) throw error;
+      this.options.onPollError?.(error.cause);
+    }
+  }
+}
+
+function wrapFetcher(fetcher: Fetcher): Fetcher {
+  return async (url) => {
+    try {
+      return await fetcher(url);
+    } catch (error) {
+      throw new FetchError(error);
+    }
+  };
+}
+
+function pollIntervalFromEnv(): number {
+  const value = process.env.POLL_INTERVAL_MS;
+  if (value === undefined) return DEFAULT_POLL_INTERVAL_MS;
+  const interval = Number(value);
+  if (!Number.isFinite(interval) || interval < 0) {
+    throw new Error('POLL_INTERVAL_MS must be a non-negative number');
+  }
+  return interval;
+}
+
+function delay(ms: number, signal?: AbortSignal): Promise<void> {
+  if (signal?.aborted) return Promise.resolve();
+  return new Promise((resolve) => {
+    const timer = setTimeout(done, ms);
+    signal?.addEventListener('abort', done, { once: true });
+
+    function done(): void {
+      clearTimeout(timer);
+      signal?.removeEventListener('abort', done);
+      resolve();
+    }
+  });
+}

sdk/src/index.ts

@@ -113,6 +113,7 @@ export { JOURNAL_WRITE_FAILED, PROTOCOL_VERSION } from './protocol.js';
 
 export { JournalClient, type JournalClientOptions } from './journal-client.js';
 export { AgentWorker, type AgentWorkerOptions } from './worker.js';
+export { HnMonitorRunner, type HnMonitorRunnerOptions } from './hn-monitor-runner.js';
 
 export {
   validateWorkPackage,

sdk/src/worker.ts

@@ -39,6 +39,7 @@ export class AgentWorker extends EventEmitter {
   }
 
   close(): void {
+    // Protocol v0 has no worker release verb; close only detaches dispatch handling.
     this.client.off('step.dispatch', this.onDispatch);
     this.attached = false;
   }