Upgrade pi to 0.83.0 and adopt context-first harness tools

[rgarcia]

Summary

Upgrades the whole pi stack to exact 0.83.0 (@earendil-works/pi-ai, pi-agent-core, pi-coding-agent, pi-tui) with one deduplicated instance of each, and adopts pi 0.83's context-first harness API as a breaking release: @onkernel/cua-ai 0.10.0, @onkernel/cua-agent 0.10.0, @onkernel/cua-cli 0.9.0 (0.8.0 shipped from main with queued-turn steering while this PR was open).

Harness redesign (@onkernel/cua-agent)

• CuaAgentHarness and CuaAgentHarnessOptions are now TContext-first — CuaAgentHarness<TContext, TSkill, TPromptTemplate> — mirroring pi's AgentHarness generic order and semantics. The supplied toolContext is forwarded to pi untouched.
• Executable harness tools are pi AgentHarnessTool<TContext> via the new CuaHarnessTool<TContext> union (a CUA spec or an AgentHarnessTool). CuaAgent stays on the ordinary pi AgentTool (CuaAgentTool); the two tool APIs are no longer conflated. CuaToolManager now materializes both views from one compiled catalog.
• Removed CuaAgentHarnessOptions.env, CuaAgentHarness.env, and CuaSystemPromptCallback — no aliases. Execution environments travel through the tool context; systemPrompt is pi's AgentHarnessSystemPrompt.
• CuaAgentHarnessOptions.retry omits pi's new harness retry?: RetryPolicy (compaction/branch-summary) so it keeps meaning CUA's provider retry policy, unchanged.
• streamFn stays optional on CuaAgentOptions (CUA supplies its default stream) even though pi 0.83 makes AgentOptions.streamFn required.
• Published declarations target pi's TypeBox 1.3 as-is.

CLI (@onkernel/cua-cli)

• Coding tools moved from pi-coding-agent createCodingTools(cwd) to pi-agent-core createReadTool/createBashTool/createEditTool/createWriteTool (read/bash/edit/write order preserved), with toolContext: { env: new NodeExecutionEnv({ cwd }) }.
• Kimi K3 now follows pi's catalog metadata with no override: low/high/max map through thinkingLevelMap, and the CLI default (--thinking unset → low) sends reasoning_effort: "low" to Moonshot and reasoning: { effort: "low" } through OpenRouter. New payload regression tests pin both.

Models (@onkernel/cua-ai)

• Removed the local claude-opus-5, gemini-3.6-flash, and gemini-3.5-flash-lite overrides — pi 0.83's registry carries all three with the same metadata. Overrides remain only for the CUA-only providers pi does not ship (Meta, Tzafon, Yutori).

Tzafon native action-loop guard (@onkernel/cua-ai)

• Tzafon's native continuation protocol requires every computer_call_output to carry an image. CUA does not synthesize post-action screenshots, so native non-screenshot actions now fail before browser execution instead of entering an unsupported text-only loop. Explicit screenshot and terminal answer actions remain supported.
• Explicit Tzafon screenshot results are retained in model context even beyond toolResultImageReplayLimit, since pruning them would make valid history impossible to replay through that protocol. Other tool-result images remain bounded.

Tests

• npm run typecheck (clean build): pass.
• @onkernel/cua-ai: 114 passed. @onkernel/cua-agent: 291 passed, 19 skipped (live tests self-skip without KERNEL_API_KEY). @onkernel/cua-cli: 141 passed, 14 skipped (ptywright native binding not built locally).
• New: harness-context.test.ts proves a custom harness tool receives the exact supplied context object and drives pi's native read/write/edit/bash tools through CUA against a real temp directory; published-declarations.test.ts compiles a downstream consumer against the packaged dist/ declarations with skipLibCheck: false; kimi-reasoning-payload.test.ts pins the default-low payloads; Tzafon provider tests pin the native non-screenshot rejection before browser execution.
• Built CLI smoke-tested (cua models, cua --help) without secrets.
• npm ls: single 0.83.0 of each pi package, single typebox 1.3.7.
• npm audit: 3 remaining advisories (brace-expansion, undici via pi-coding-agent, and a moderate) are pre-existing on main and pinned by upstream dependency ranges; not changed by this PR.
• Pre-existing, unrelated: packages/ptywright's own test suite fails on main with extensionless ESM imports (ERR_MODULE_NOT_FOUND for dist/index); this PR does not touch ptywright. Latent type errors in browser-wait.test.ts / cli-executor.test.ts also predate this change (test files are outside every tsconfig project).

---

Note

High Risk
Breaking public harness and CLI APIs plus a major pi/TypeBox bump affect all consumers; Tzafon and image-replay behavior changes are easy to miss in integration testing.

Overview
Bumps the monorepo to pi 0.83.0 (pi-ai, pi-agent-core, pi-coding-agent, pi-tui) and ships breaking 0.10.0 agent/ai and 0.9.0 CLI releases.

CuaAgentHarness is now context-first: generics are CuaAgentHarness<TContext, …>, harness tools are CuaHarnessTool / pi AgentHarnessTool with toolContext forwarded on each call, and env on the harness is removed (execution env goes through toolContext, e.g. NodeExecutionEnv for read/bash/edit/write). CuaAgent keeps context-free CuaAgentTool; CuaToolManager exposes both agent and harness tool views from one catalog.

CLI swaps createCodingTools for pi-agent-core read/bash/edit/write tools and wires toolContext: { env }. Kimi K3 default thinking maps to reasoning_effort: low per pi catalog metadata.

@onkernel/cua-ai drops local overrides for Opus 5 and Gemini models now in pi’s registry. Tzafon rejects native non-screenshot computer actions and text-only native tool results because continuation requires images CUA does not auto-capture. Agent exempts Tzafon native screenshot tool results from toolResultImageReplayLimit trimming.

New tests cover harness tool context, published declarations with skipLibCheck: false, Kimi payloads, and Tzafon guards; docs note pi 0.83.0 for dynamic setTools() behavior.

^{Reviewed by Cursor Bugbot for commit e922447. Bugbot is set up for automated code reviews on this repo. Configure here.}

[cursor]

Cursor Bugbot has reviewed your changes using high effort and found 1 potential issue.

There are 2 total unresolved issues (including 1 from previous review).

Autofix Details

Bugbot Autofix prepared a fix for the issue found in the latest run.

• ✅ Fixed: Missing TSDoc on exported helper
  • Added a TSDoc block to the exported mapThinkingLevel helper documenting purpose, defaulting behavior, and invalid-input error contract.

Or push these changes by commenting:

@cursor push 21209a2a93

Preview (21209a2a93)

diff --git a/packages/cli/src/cli-harness.ts b/packages/cli/src/cli-harness.ts
--- a/packages/cli/src/cli-harness.ts
+++ b/packages/cli/src/cli-harness.ts
@@ -487,6 +487,12 @@
 	return value && value.length > 0 ? value : undefined;
 }
 
+/**
+ * Normalize CLI `--thinking` values to canonical harness thinking levels.
+ *
+ * Defaults to `"low"` when the flag is unset or empty, and throws when the
+ * input does not match a supported level.
+ */
 export function mapThinkingLevel(raw: string | undefined): "off" | "minimal" | "low" | "medium" | "high" | "xhigh" | "max" {
 	const v = (raw ?? "low").trim().toLowerCase();
 	switch (v) {

_{You can send follow-ups to the cloud agent here.}

[cursor]

Cursor Bugbot has reviewed your changes using high effort and found 1 potential issue.

Bugbot Autofix prepared a fix for the issue found in the latest run.

• ✅ Fixed: Tzafon image guard blocks continuations
  • Native Tzafon tool-result replay now emits a text-backed computer_screenshot.error payload when no image is present instead of throwing, so valid text-only continuations no longer hard-fail.

Or push these changes by commenting:

@cursor push 62940dda82

Preview (62940dda82)

diff --git a/packages/ai/src/providers/tzafon/provider.ts b/packages/ai/src/providers/tzafon/provider.ts
--- a/packages/ai/src/providers/tzafon/provider.ts
+++ b/packages/ai/src/providers/tzafon/provider.ts
@@ -344,15 +344,12 @@
 				.trim();
 			const image = [...message.content].reverse().find((part): part is ImageContent => part.type === "image");
 			if (nativeComputerName && message.toolName === nativeComputerName) {
-				if (!image) {
-					throw new Error(
-						"Tzafon native computer action loops require image tool results; text-only results are unsupported because CUA does not capture post-action screenshots automatically.",
-					);
-				}
 				items.push({
 					type: "computer_call_output",
 					call_id: message.toolCallId,
-					output: { type: "computer_screenshot", image_url: `data:${image.mimeType};base64,${image.data}` },
+					output: image
+						? { type: "computer_screenshot", image_url: `data:${image.mimeType};base64,${image.data}` }
+						: { type: "computer_screenshot", error: message.isError ? text || "tool execution failed" : text || "no screenshot" },
 				});
 				continue;
 			}

diff --git a/packages/ai/test/tzafon-provider.test.ts b/packages/ai/test/tzafon-provider.test.ts
--- a/packages/ai/test/tzafon-provider.test.ts
+++ b/packages/ai/test/tzafon-provider.test.ts
@@ -99,8 +99,8 @@
 		expect(toolCalls(message.content)).toEqual([]);
 	});
 
-	it("rejects text-only native computer results before sending a request", () => {
-		expect(() => tzafon.buildTzafonRequestInput(model, {
+	it("serializes text-only native computer results as computer_screenshot errors", () => {
+		const payload = tzafon.buildTzafonRequestInput(model, {
 			messages: [
 				{
 					role: "assistant",
@@ -125,7 +125,19 @@
 		}, {
 			disableResponseThreading: true,
 			cuaIncomingToolPlan: { tzafonComputerName: "computer", yutoriNames: {}, googleNames: {}, googleExcludedNames: [], nativeToolNames: ["computer"] },
-		})).toThrow("text-only results are unsupported");
+		});
+
+		expect(payload.input).toEqual(expect.arrayContaining([
+			expect.objectContaining({ type: "computer_call", call_id: "call_click" }),
+			expect.objectContaining({
+				type: "computer_call_output",
+				call_id: "call_click",
+				output: {
+					type: "computer_screenshot",
+					error: "Actions executed successfully.",
+				},
+			}),
+		]));
 	});
 
 	it("degrades malformed function-call arguments to empty args instead of failing the turn", async () => {

_{You can send follow-ups to the cloud agent here.}

^{Reviewed by Cursor Bugbot for commit f588ba4. Configure here.}