localpi 0.5.0 → 0.5.2

package/README.md

@@ -18,6 +18,12 @@ See:
 npm install -g localpi
 ```
 
+Or the latest GitHub release directly:
+
+```bash
+npm install -g https://github.com/osolmaz/localpi/releases/latest/download/localpi.tgz
+```
+
 During development:
 
 ```bash
@@ -74,6 +80,14 @@ Localpi launches Pi with:
 
 The approval gate makes failed or denied tool calls explicit to the model so the model does not claim that a blocked command ran.
 
+## Diffusion Canvas Visualizer
+
+The live diffusion canvas visualizer (watching DiffusionGemma denoise its
+answer in the TUI) lives in its own project now:
+[diffusionpi](https://github.com/osolmaz/diffusionpi). It is a pi-factory app
+bundle plus a standalone Pi widget package; the widget also installs into any
+Pi session via `pi install`.
+
 ## LM Studio Alternative
 
 LM Studio exposes an OpenAI-compatible endpoint, usually:
@@ -115,7 +129,7 @@ Run an endless TUI demo:
 localpi --demo --model gemma-e4b
 ```
 
-Demo mode requires an explicit model, opens the normal Pi TUI, and keeps one live Pi session so followup prompts continue from the first prompt while Pi owns streaming, tok/s status, slash commands, and exit behavior.
+Demo mode requires an explicit model, opens the normal Pi TUI, and keeps one live Pi session so followup prompts continue from the first prompt while Pi owns streaming, tok/s status, slash commands, and exit behavior. Under the hood it loads the shared [pi-demo-mode](https://github.com/osolmaz/pi-demo-mode) extension, configured through the `--demo-*` flags below.
 
 Override the demo prompts:
 
@@ -174,6 +188,18 @@ Stop the managed `llama-server` runtime:
 localpi --stop
 ```
 
+## Demo Grid and Recording
+
+The `localpi grid` and `localpi record` subcommands moved to
+[demowall](https://github.com/osolmaz/demowall), a standalone tool that runs
+N copies of any command in a tiled tmux wall and records tmux sessions to
+video. A wall of localpi demo sessions is:
+
+```bash
+demowall grid --concurrency 4 --start -- localpi --demo --model gemma-e4b
+demowall record --session demowall-<timestamp> --out demo.mp4 --seconds 60
+```
+
 ## Options
 
 - `--runtime <auto|llama-server|lmstudio|vllm|openai-compatible>`: runtime backend. Default: `auto`

package/dist/src/cli/cli.js

@@ -1,4 +1,4 @@
-import { runPiApp } from "@dutifuldev/pi-factory";
+import { runPiApp } from "@osolmaz/pi-factory";
 import { errorMessage, fail, ok } from "../common/result.js";
 import { parseLocalpiArgs, usage } from "../localpi/options.js";
 import { aliasListOutput, connectionStatus, resolveRuntime, statusOutput, stopRuntime } from "../localpi/runtime.js";
@@ -22,7 +22,9 @@ export async function run(args) {
         });
         const connection = await resolveRuntime(options);
         const selectorOptions = startupModelSelectorOptions(options, connection);
-        const extensions = await writeDefaultExtensions(options, selectorOptions === undefined ? {} : { startupModelSelector: selectorOptions });
+        const extensions = await writeDefaultExtensions(options, {
+            ...(selectorOptions === undefined ? {} : { startupModelSelector: selectorOptions })
+        });
         const app = createLocalpiAppDefinition(options, connection, extensions);
         return await launchResolvedRuntime(app, connection);
     }

package/dist/src/pi/app.js

@@ -143,6 +143,7 @@ function extensionConfig(extensions) {
     }
     return {
         extensions: extensions.paths.map((extensionPath) => ({ path: extensionPath })),
-        appendSystemPrompts: [extensions.systemPrompt]
+        appendSystemPrompts: [extensions.systemPrompt],
+        ...(Object.keys(extensions.env).length === 0 ? {} : { env: extensions.env })
     };
 }

package/dist/src/pi/extension-sources/token-status.js

@@ -1,6 +1,9 @@
-export function tokenStatusExtensionSource() {
+export function tokenStatusExtensionSource(options = {}) {
+    const includeContext = options.includeContext ?? true;
     return `import type { ExtensionAPI } from "@earendil-works/pi-coding-agent";
 
+const includeContext = ${JSON.stringify(includeContext)};
+
 type Usage = {
   input?: number;
   output?: number;
@@ -74,9 +77,10 @@ export default function localpiTokenStatus(pi: ExtensionAPI): void {
     const elapsedSeconds = elapsed(state, now);
     const decodeSeconds = generationElapsed(state, now);
     const prefillText = prefillStatusText(state, input, cacheWrite);
-    const context = ctx.getContextUsage();
-    const contextText =
-      context && context.percent !== null
+    const context = includeContext ? ctx.getContextUsage() : undefined;
+    const contextText = !includeContext
+      ? undefined
+      : context && context.percent !== null
         ? \`ctx \${Math.round(context.percent)}%/\${Math.round(context.contextWindow / 1000)}k\`
         : "ctx ?";
 

package/dist/src/pi/extensions.js

@@ -1,8 +1,8 @@
 import { mkdir, writeFile } from "node:fs/promises";
+import { createRequire } from "node:module";
 import path from "node:path";
 import { localpiSettingsPath } from "../localpi/settings-state.js";
 import { resolveDemoPrompts } from "./demo.js";
-import { demoModeExtensionSource } from "./extension-sources/demo-mode.js";
 import { startupModelSelectorExtensionSource } from "./extension-sources/startup-model-selector.js";
 import { thinkingControlExtensionSource } from "./extension-sources/thinking-control.js";
 import { tokenStatusExtensionSource } from "./extension-sources/token-status.js";
@@ -11,24 +11,40 @@ export async function writeDefaultExtensions(options, extensionOptions = {}) {
     const extensionDir = path.join(options.stateDir, "pi-extensions");
     await mkdir(extensionDir, { recursive: true });
     const paths = [];
+    let env = {};
     if (extensionOptions.startupModelSelector !== undefined) {
         paths.push(await writeExtension(extensionDir, "startup-model-selector.ts", startupModelSelectorExtensionSource(extensionOptions.startupModelSelector)));
     }
     if (options.demo) {
-        paths.push(await writeExtension(extensionDir, "demo-mode.ts", demoModeExtensionSource(await resolveDemoPrompts(options))));
+        paths.push(demoModeExtensionPath());
+        const prompts = await resolveDemoPrompts(options);
+        env = {
+            ...env,
+            PI_DEMO_MODE: "1",
+            PI_DEMO_INITIAL_PROMPT: prompts.initial,
+            PI_DEMO_FOLLOWUP_PROMPT: prompts.followup
+        };
     }
     paths.push(await writeExtension(extensionDir, "thinking-control.ts", thinkingControlExtensionSource(localpiSettingsPath(options))));
     if (options.approval) {
         paths.push(await writeExtension(extensionDir, "tool-approval.ts", approvalExtensionSource()));
     }
     if (options.tokenStatus) {
-        paths.push(await writeExtension(extensionDir, "token-status.ts", tokenStatusExtensionSource()));
+        paths.push(await writeExtension(extensionDir, "token-status.ts", tokenStatusExtensionSource({ includeContext: !options.demo })));
     }
     return {
         paths,
+        env,
         systemPrompt: localpiSystemPrompt(options.approval)
     };
 }
+// Demo mode is the shared pi-demo-mode package (a git dependency), loaded
+// straight from node_modules and configured through PI_DEMO_* env vars.
+export function demoModeExtensionPath() {
+    const require = createRequire(import.meta.url);
+    const packageJson = require.resolve("pi-demo-mode/package.json");
+    return path.join(path.dirname(packageJson), "extensions", "demo-mode.ts");
+}
 async function writeExtension(extensionDir, name, source) {
     const extensionPath = path.join(extensionDir, name);
     await writeFile(extensionPath, source, "utf8");

package/package.json

@@ -1,17 +1,17 @@
 {
   "name": "localpi",
-  "version": "0.5.0",
+  "version": "0.5.2",
   "description": "Swiss army knife for running Pi with local inference engines.",
   "type": "module",
   "license": "MIT",
   "repository": {
     "type": "git",
-    "url": "git+https://github.com/dutifuldev/localpi.git"
+    "url": "git+https://github.com/osolmaz/localpi.git"
   },
   "bugs": {
-    "url": "https://github.com/dutifuldev/localpi/issues"
+    "url": "https://github.com/osolmaz/localpi/issues"
   },
-  "homepage": "https://github.com/dutifuldev/localpi#readme",
+  "homepage": "https://github.com/osolmaz/localpi#readme",
   "publishConfig": {
     "access": "public"
   },
@@ -26,7 +26,8 @@
     "LICENSE"
   ],
   "scripts": {
-    "build": "tsc -p tsconfig.json",
+    "build": "tsc -p tsconfig.json && chmod +x dist/src/cli/main.js",
+    "prepare": "npm run build",
     "format": "prettier --check .",
     "lint": "eslint .",
     "typecheck": "tsc --noEmit",
@@ -48,13 +49,14 @@
     "@vitest/coverage-v8": "^3.0.0",
     "eslint": "^9.0.0",
     "prettier": "^3.0.0",
-    "slophammer-ts": "0.4.0",
+    "slophammer-ts": "0.4.1",
     "tsx": "^4.19.4",
     "typescript": "^5.0.0",
     "typescript-eslint": "^8.0.0",
     "vitest": "^3.0.0"
   },
   "dependencies": {
-    "@dutifuldev/pi-factory": "^0.1.1"
+    "@osolmaz/pi-factory": "^0.1.2",
+    "pi-demo-mode": "github:osolmaz/pi-demo-mode#v0.1.0"
   }
 }

package/dist/src/pi/extension-sources/demo-mode.js

@@ -1,110 +0,0 @@
-export function demoModeExtensionSource(prompts) {
-    const initialPromptSource = JSON.stringify(prompts.initial);
-    const followupPromptSource = JSON.stringify(prompts.followup);
-    return `import type { ExtensionAPI, ExtensionContext, TurnEndEvent } from "@earendil-works/pi-coding-agent";
-
-const initialPrompt = ${initialPromptSource};
-const followupPrompt = ${followupPromptSource};
-const compactAtContextPercent = 70;
-const demoCompactionInstructions = [
-  "Preserve the demo narrative state, named entities, current setting,",
-  "unresolved plot threads, and latest user direction.",
-  "Keep the summary concise so the story can continue after compaction."
-].join(" ");
-
-export default function localpiDemoMode(pi: ExtensionAPI): void {
-  let started = false;
-  let stopped = false;
-  let compacting = false;
-
-  function queueInitialPrompt(): void {
-    queueMicrotask(() => {
-      if (!stopped) {
-        pi.sendUserMessage(initialPrompt);
-      }
-    });
-  }
-
-  function queueFollowup(): void {
-    queueMicrotask(() => {
-      if (!stopped && !compacting) {
-        pi.sendUserMessage(followupPrompt, { deliverAs: "followUp" });
-      }
-    });
-  }
-
-  function compactThenFollowup(ctx: ExtensionContext): void {
-    if (compacting) {
-      return;
-    }
-    compacting = true;
-    ctx.compact({
-      customInstructions: demoCompactionInstructions,
-      onComplete: () => {
-        compacting = false;
-        queueFollowup();
-      },
-      onError: (error) => {
-        compacting = false;
-        stopped = true;
-        ctx.ui.notify("Demo compaction failed: " + error.message, "error");
-      }
-    });
-  }
-
-  pi.on("session_start", (event, ctx) => {
-    if (started || stopped || event.reason !== "startup" || ctx.mode !== "tui") {
-      return;
-    }
-    started = true;
-    queueInitialPrompt();
-  });
-
-  pi.on("turn_end", (event, ctx) => {
-    if (!started || stopped || compacting || ctx.mode !== "tui") {
-      return;
-    }
-    if (event.message.role !== "assistant") {
-      return;
-    }
-    switch (event.message.stopReason) {
-      case "aborted":
-      case "error":
-        stopped = true;
-        return;
-      case "toolUse":
-        return;
-    }
-    if (shouldCompactBeforeFollowup(event, ctx)) {
-      compactThenFollowup(ctx);
-      return;
-    }
-    queueFollowup();
-  });
-
-  pi.on("session_shutdown", () => {
-    stopped = true;
-  });
-}
-
-function shouldCompactBeforeFollowup(event: TurnEndEvent, ctx: ExtensionContext): boolean {
-  const contextPercent = currentContextPercent(event, ctx);
-  return contextPercent !== undefined && contextPercent >= compactAtContextPercent;
-}
-
-function currentContextPercent(event: TurnEndEvent, ctx: ExtensionContext): number | undefined {
-  const usage = ctx.getContextUsage();
-  if (usage?.percent !== undefined && usage.percent !== null) {
-    return usage.percent;
-  }
-  if (event.message.role !== "assistant") {
-    return undefined;
-  }
-  const contextWindow = ctx.model?.contextWindow;
-  if (contextWindow === undefined || contextWindow <= 0) {
-    return undefined;
-  }
-  return (event.message.usage.totalTokens / contextWindow) * 100;
-}
-`;
-}