// AI Platform Drift

Cloudflare Agents SDK v0.33.0 upgrades to Pi 1.5 with breaking API changes

Cloudflare developer changelog · published 2026-11-11 05:19 UTC
ACTION NEEDED MIGRATE APIDEPRECATION
WHAT CHANGEDAgents SDK v0.33.0 upgrades PiHarness to Pi 1.5. Breaking changes: session.events() replaced by session.watch(), retry() method removed, parked status removed. New APIs: session.continue() for resuming conversations without adding a message, defaults.maxTurns parameter for capping tool rounds. Once a conversation runs on Pi 1.5, Pi 1.4 cannot read it; rollback to Pi 1.4 is not supported. Performance improvements: 27x fewer storage writes for long answers, 5.7x for parallel tools, throughput increased from 88 to 459 agent runs per second in a 2,048-session benchmark. Crash recovery window: up to about one second of streamed progress can be lost with default settings.
WHY IT MATTERSCode calling session.events(), retry(), or checking for parked status will break. Deployments cannot roll back to Pi 1.4 after upgrading because Pi 1.5 sessions are unreadable by Pi 1.4. A crash during streaming can now lose up to about one second of progress instead of none.
WHAT TO DOUpdate all callers of session.events() to use session.watch() and remove uses of retry() and the parked status. Review the Pi Durable 1.5.0 changelog and Agents SDK v0.33.0 changelog for the complete list of breaking changes and step-by-step upgrade instructions. Plan to roll forward only if deployment issues occur after upgrading.
Model-reported confidence 95% (not measured accuracy) Primary source ↗ written by anthropic/claude-sonnet-4.5
Show the reasoning — raw diff, triage verdict, confidence
RAW DIFF
+## Agents - Agents SDK now supports Pi 1.5 with fewer storage writes and lower memory use
+https://developers.cloudflare.com/changelog/post/2026-11-03-pi-harness-pi-1-5/
+Tue, 03 Nov 2026 00:00:00 GMT
+
+Agents SDK v0.33.0 upgrades PiHarness to Pi 1.5 ↗︎. The release reduces storage transactions while streaming, improves throughput across many sessions, and reduces memory use when streaming large tool calls.
+Performance improvements
+
+Far fewer storage writes while streaming. Pi stores streamed answers and tool output in batches, within a second of each update, instead of on every update. Over SQLite, this cuts storage transactions by about 27x for long answers and 5.7x for turns with parallel tools, and CPU by about a quarter.
+Higher throughput across many sessions. Pi's scheduler no longer rechecks every conversation on each commit. In a SQLite benchmark with 2,048 sessions, one Pi harness went from 88 to 459 agent runs per second.
+Less memory on large answers. Streaming a 500 KB tool call used to take about 20 seconds of CPU and a 100 MB heap. Pi now appends new text instead of copying it on every update.
+
+Streaming updates remain visible immediately, but are persisted in batches. With the default settings, a crash can lose up to about one second of streamed progress. Refer to Recovery to configure this interval.
+Session and tool updates
+
+A view of each session. session.watch() replaces session.events(). It returns the newest transcript entries and the session's live state, such as the answer being streamed, running tools, and why the last run ended without an answer.
+Continue without adding a message. The new session.continue() API starts a new run using the existing transcript, for example after a model error or when a run reaches its turn limit.
+Turn limit. defaults.maxTurns caps the tool rounds a run can take in new sessions.
+Tasks that keep crashing end. Pi now ends a task cut off by its restart limit instead of parking it. retry() and the parked status are removed.
+Browser in Pi's code mode. Scripts in Pi's code mode can call browserTool and pass its screenshots to image(). browserTool also stops when the harness aborts its call.
+
+Continue an unanswered conversation
+For an existing session and submission operationId, check the result before offering a continuation. For example, after resolving a model error:
+const result = await session.wait(operationId);
+
+if (result.status === "unanswered" && result.reason === "model_error") {
+const receipt = await session.continue({
+operationId: `${operationId}:continue`,
+});
+const continued = await session.wait(receipt.operationId);
+console.log(continued.status, continued.text);
+}const result = await session.wait(operationId);
+
+if (result.status === "unanswered" && result.reason === "model_error") {
+const receipt = await session.continue({
+operationId: `${operationId}:continue`,
+});
+const continued = await session.wait(receipt.operationId);
+console.log(continued.status, continued.text);
+}
+This starts a new run without appending a user message. The session must be idle. If the transcript already ends in an answer, the operation returns unanswered with reason nothing_to_answer. Refer to Work with sessions.
+Upgrade
+Once a conversation runs on Pi 1.5, Pi 1.4 cannot read it. After upgrading, roll forward to fix deployment issues rather than rolling back to Pi 1.4.
+Update callers of session.events() to use session.watch(), and remove uses of retry() and the parked status. Refer to the Pi Durable 1.5.0 changelog ↗︎ for the full list of breaking changes.
+The Agents SDK v0.33.0 changelog ↗︎ has step-by-step upgrade instructions. Point your coding agent at them and upgrade today:
+Copy promptPrompt copied!
HOW IT WAS JUDGED
Triage signals below are model-reported assertions, preserved verbatim; they are not independent arithmetic verification.
triage verdictMATERIAL at 0.92, needed 0.5
triage reasoningBreaking API changes, new session management APIs, and one-way upgrade path require developer action.
signalsbreaking API changes: session.events() → session.watch(), retry() removed, parked status removed · new API: session.continue() for resuming conversations · new parameter: defaults.maxTurns for turn limits · backward incompatibility: Pi 1.5 sessions unreadable by Pi 1.4 · upgrade requirement: roll forward only, no rollback path · performance improvements quantified: 27x fewer storage writes, 5.7x for parallel tools, 459 vs 88 runs/sec · behavioral change: crash recovery window up to ~1 second of streamed progress loss
suppression—
triage modelanthropic/claude-haiku-4.5 ($0.00000)
writer modelanthropic/claude-sonnet-4.5 ($0.00000)