pi-ui-extend 0.1.24 → 0.1.25

package/dist/app/input/autocomplete-controller.js

@@ -177,7 +177,6 @@ export async function completeInputWithPi(runtime, draft, config, signal) {
             maxTokens: requestModel.maxTokens,
             ...(parsedModel.thinkingLevel && parsedModel.thinkingLevel !== "off" ? { reasoning: parsedModel.thinkingLevel } : {}),
             signal: requestSignal.signal,
-            temperature: 0.1,
             timeoutMs,
         });
         for await (const event of stream) {

package/dist/schemas/pi-tools-suite-schema.d.ts

@@ -83,7 +83,6 @@ export declare const PiToolsSuiteConfigSchema: Type.TObject<{
             maxTaskChars: Type.TOptional<Type.TNumber>;
             maxTokens: Type.TOptional<Type.TNumber>;
             maxRetries: Type.TOptional<Type.TNumber>;
-            temperature: Type.TOptional<Type.TNumber>;
             timeoutMs: Type.TOptional<Type.TNumber>;
             debug: Type.TOptional<Type.TBoolean>;
         }>>;

package/dist/schemas/pi-tools-suite-schema.js

@@ -115,7 +115,6 @@ const SubagentRoutingConfig = Type.Object({
     maxTaskChars: Type.Optional(Type.Number({ description: "Max task/scope characters sent to router.", minimum: 100 })),
     maxTokens: Type.Optional(Type.Number({ description: "Max router response tokens.", minimum: 8 })),
     maxRetries: Type.Optional(Type.Number({ description: "Router request retries.", minimum: 0 })),
-    temperature: Type.Optional(Type.Number({ description: "Router sampling temperature.", minimum: 0, maximum: 2 })),
     timeoutMs: Type.Optional(Type.Number({ description: "Router request timeout in ms.", minimum: 1000 })),
     debug: Type.Optional(Type.Boolean({ description: "Show routing debug warnings." })),
 }, { description: "LLM-based role routing configuration." });

package/extensions/session-title/config.ts

@@ -12,7 +12,6 @@ export interface SessionTitleConfig {
 	maxRetries: number;
 	generationAttempts: number;
 	retryDelayMs: number;
-	temperature: number;
 	timeoutMs: number;
 	terminalTitle: boolean;
 	terminalTitlePrefix: string;
@@ -29,7 +28,6 @@ const DEFAULT_CONFIG: SessionTitleConfig = {
 	maxRetries: 2,
 	generationAttempts: 3,
 	retryDelayMs: 3000,
-	temperature: 0.2,
 	timeoutMs: 12_000,
 	terminalTitle: true,
 	terminalTitlePrefix: "pi — ",
@@ -83,9 +81,6 @@ function mergeConfig(base: SessionTitleConfig, raw: Record<string, unknown>): Se
 	if (typeof raw.retryDelayMs === "number" && Number.isFinite(raw.retryDelayMs)) {
 		next.retryDelayMs = Math.max(250, Math.floor(raw.retryDelayMs));
 	}
-	if (typeof raw.temperature === "number" && Number.isFinite(raw.temperature)) {
-		next.temperature = Math.min(2, Math.max(0, raw.temperature));
-	}
 	if (typeof raw.timeoutMs === "number" && Number.isFinite(raw.timeoutMs)) {
 		next.timeoutMs = Math.max(1000, Math.floor(raw.timeoutMs));
 	}

package/extensions/session-title/index.ts

@@ -221,7 +221,6 @@ async function generateSessionTitle(
 			maxRetries: config.maxRetries,
 			maxTokens: config.maxTokens,
 			signal,
-			temperature: config.temperature,
 			timeoutMs: config.timeoutMs,
 		},
 	);

package/external/pi-tools-suite/README.md

@@ -163,7 +163,7 @@ Async-subagents also injects a lightweight oh-my-openagent-style system-prompt s
 
 When the parent model cannot inspect images, async-subagents adds vision-delegation guidance and can save current-turn image attachments under `.pi/subagents/attachments/` so a `vision` sub-agent can receive them as `imagePaths`. Dynamic provider capabilities can be missing or stale after switching models, so blind parent models can still be configured explicitly with case-insensitive `*` masks under `asyncSubagents.vision.blindModelPatterns` in `~/.config/pi/pi-tools-suite.jsonc`. GLM is no longer treated as blind by async-subagents by default; the main-session `glm-coding-discipline` lookup tool is the preferred path for GLM visual lookups.
 
-When a task omits `subagentType`, async-subagents asks a lightweight router model to choose one configured type for each task from the task text/scope and the `types.<name>.description` metadata. Explicit task `subagentType` still wins. Keep type descriptions short, literal, and distinct because they are inserted into the router prompt for a small model. Router settings live under `asyncSubagents.routing` (`enabled`, `model`, `maxTaskChars`, `maxTokens`, `maxRetries`, `temperature`, `timeoutMs`, `debug`); the default router model is `zai/glm-4.5-air`. If the router is disabled, unavailable, aborted, or returns invalid JSON, omitted types fall back to `defaultType`.
+When a task omits `subagentType`, async-subagents asks a lightweight router model to choose one configured type for each task from the task text/scope and the `types.<name>.description` metadata. Explicit task `subagentType` still wins. Keep type descriptions short, literal, and distinct because they are inserted into the router prompt for a small model. Router settings live under `asyncSubagents.routing` (`enabled`, `model`, `maxTaskChars`, `maxTokens`, `maxRetries`, `timeoutMs`, `debug`); the default router model is `zai/glm-4.5-air`. If the router is disabled, unavailable, aborted, or returns invalid JSON, omitted types fall back to `defaultType`.
 
 Define optional `presets` under `asyncSubagents` in `~/.config/pi/pi-tools-suite.jsonc`, `$PI_CONFIG_DIR/pi-tools-suite.jsonc`, or project `.pi/pi-tools-suite.jsonc`, then use `/subagent-preset` or `/subagent-preset-config` to pick one persistent active preset for future spawns across all sessions. Set `AGENTS_PRESET=<name>` before launching Pi to override the saved preset for only the current process/session without changing the saved selection. If Pi is already running, use `/subagent-preset session <name>` for the same process-only override, and `/subagent-preset session-clear` to remove that runtime override. The TUI only selects presets already present in config; it does not edit JSON. If no `asyncSubagents` section exists, run `/subagent-preset init` to insert the bundled sample from `src/async-subagents/async-subagents.sample.jsonc` into the shared config (or to copy a standalone override file when `ASYNC_SUBAGENTS_CONFIG` / `PI_SUBAGENTS_CONFIG` is set). Existing config sections/files are never overwritten. Presets select an agent/model configuration: they can provide global fallback `model`/`thinking`/`extraArgs` and per-role overrides under `asyncSubagents.presets.<name>.types.<subagentType>`. They can also provide ordered `fallbackModels` globally or per-role; when a sub-agent fails with quota/rate-limit errors such as 429, async-subagents immediately tries the next fallback model and remembers the exhausted provider for the current Pi process/session, so later spawns skip that provider until Pi exits. This is intended for provider-level fallback chains such as `antigravity/* → openai-codex/* → zai/*` or `openai-codex/* → zai/*`; omit fallbacks for effectively unlimited providers. Antigravity account rotation has priority over preset fallback: async-subagents only falls back after Antigravity reports that all configured accounts are exhausted for that model. Explicit task model overrides and force-current-model disable preset fallback for that task. The active preset name is stored separately in `~/.pi/agent/subagent-preset-selection.json`.
 

package/external/pi-tools-suite/src/async-subagents/async-subagents.sample.jsonc

@@ -58,7 +58,6 @@
     "maxTaskChars": 1200,
     "maxTokens": 512,
     "maxRetries": 1,
-    "temperature": 0,
     "timeoutMs": 12000,
     "debug": false
   },

package/external/pi-tools-suite/src/async-subagents/core/config.ts

@@ -37,8 +37,6 @@ export interface SubagentRoutingConfig {
 	maxTokens?: number;
 	/** Router complete() retries. */
 	maxRetries?: number;
-	/** Router sampling temperature. */
-	temperature?: number;
 	/** Router request timeout. */
 	timeoutMs?: number;
 	/** Show best-effort UI warnings when routing falls back. */
@@ -150,7 +148,6 @@ export const DEFAULT_ROUTING_CONFIG: ResolvedSubagentRoutingConfig = {
 	maxTaskChars: 1200,
 	maxTokens: 512,
 	maxRetries: 1,
-	temperature: 0,
 	timeoutMs: 12_000,
 	debug: false,
 };
@@ -522,8 +519,6 @@ function normalizeRoutingConfig(value: Record<string, unknown>): SubagentRouting
 	if (maxTokens !== undefined) routing.maxTokens = Math.max(8, Math.round(maxTokens));
 	const maxRetries = finiteNumber(value.maxRetries);
 	if (maxRetries !== undefined) routing.maxRetries = Math.max(0, Math.round(maxRetries));
-	const temperature = finiteNumber(value.temperature);
-	if (temperature !== undefined) routing.temperature = Math.min(2, Math.max(0, temperature));
 	const timeoutMs = finiteNumber(value.timeoutMs);
 	if (timeoutMs !== undefined) routing.timeoutMs = Math.max(1000, Math.round(timeoutMs));
 	return routing;

package/external/pi-tools-suite/src/async-subagents/core/routing.ts

@@ -80,7 +80,6 @@ export async function routeSubagentTasks(
 				maxRetries: routing.maxRetries,
 				maxTokens: routing.maxTokens,
 				signal,
-				temperature: routing.temperature,
 				timeoutMs: routing.timeoutMs,
 			},
 		);

package/external/pi-tools-suite/src/async-subagents/core/ultrawork-auto.ts

@@ -82,7 +82,6 @@ export async function decideUltraworkAuto(
 				maxRetries: routing.maxRetries,
 				maxTokens: Math.min(routing.maxTokens, 32),
 				signal,
-				temperature: 0,
 				timeoutMs: routing.timeoutMs,
 			},
 		);

package/external/pi-tools-suite/src/default-pi-tools-suite-config.ts

@@ -23,7 +23,7 @@ export const DEFAULT_PI_TOOLS_SUITE_CONFIG_JSONC = String.raw`{
   },
   "asyncSubagents": {
     "defaultType": "quick",
-    "routing": { "enabled": true, "model": "zai/glm-4.5-air", "maxTaskChars": 1200, "maxTokens": 512, "maxRetries": 1, "temperature": 0, "timeoutMs": 12000, "debug": false },
+    "routing": { "enabled": true, "model": "zai/glm-4.5-air", "maxTaskChars": 1200, "maxTokens": 512, "maxRetries": 1, "timeoutMs": 12000, "debug": false },
     "presets": {
       "cheap": {
         "description": "Use cheap GLM/Gemini Flash models for text/code roles; keep vision on the enabled GPT vision model.",

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "pi-ui-extend",
-  "version": "0.1.24",
+  "version": "0.1.25",
   "private": false,
   "type": "module",
   "bin": {

package/schemas/pi-tools-suite.json

@@ -419,12 +419,6 @@
               "description": "Router request retries.",
               "minimum": 0
             },
-            "temperature": {
-              "type": "number",
-              "description": "Router sampling temperature.",
-              "minimum": 0,
-              "maximum": 2
-            },
             "timeoutMs": {
               "type": "number",
               "description": "Router request timeout in ms.",