@oh-my-pi/pi-coding-agent 8.5.0 → 8.8.8

package/CHANGELOG.md

@@ -2,6 +2,29 @@
 
 ## [Unreleased]
 
+## [8.8.8] - 2026-01-28
+
+### Added
+- Added `/fork` command to create a new session with the exact same state (entries and artifacts) as the current session
+
+### Changed
+- Renamed the `complete` tool to `submit_result` for subagent result submission
+
+## [8.6.0] - 2026-01-27
+
+### Added
+- Added `plan` model role for specifying the model used by the plan agent
+- Added `--plan` CLI flag and `OMP_PLAN_MODEL` environment variable for ephemeral plan model override
+- Added plan model selection in model selector UI with PLAN badge
+
+### Changed
+- Task tool subagents now execute in-process instead of using worker threads
+
+### Fixed
+- Queued skill commands as follow-ups when the agent is already streaming to avoid load failures
+- Deduplicated repeated review findings in subagent progress rendering
+- Restored MCP proxy tool timeout handling to prevent subagent hangs
+
 ## [8.5.0] - 2026-01-27
 
 ### Added

package/README.md

@@ -263,7 +263,7 @@ The agent reads, writes, and edits files, and executes commands via bash.
 | Ctrl+D                | Exit (when editor is empty)                              |
 | Ctrl+Z                | Suspend to background (use `fg` in shell to resume)      |
 | Shift+Tab             | Cycle thinking level                                     |
-| Ctrl+P / Shift+Ctrl+P | Cycle role models (slow/default/smol)                    |
+| Ctrl+P / Shift+Ctrl+P | Cycle role models (slow/default/smol/plan)               |
 | Ctrl+L                | Open model selector                                      |
 | Ctrl+O                | Toggle tool output expansion                             |
 | Ctrl+T                | Toggle todo list expansion                               |
@@ -564,7 +564,7 @@ Global `~/.omp/agent/settings.json` stores persistent preferences:
 | Setting                       | Description                                                                                                                         | Default         |
 | ----------------------------- | ----------------------------------------------------------------------------------------------------------------------------------- | --------------- |
 | `theme`                       | Color theme name                                                                                                                    | auto-detected   |
-| `modelRoles`                  | Model assignments by role (e.g., `{"default": "anthropic/claude-sonnet-4-20250514", "slow": "...", "smol": "..."}`)                  | -               |
+| `modelRoles`                  | Model assignments by role (e.g., `{"default": "...", "slow": "...", "smol": "...", "plan": "..."}`)                                  | -               |
 | `defaultThinkingLevel`        | Thinking level: `off`, `minimal`, `low`, `medium`, `high`, `xhigh`                                                                  | -               |
 | `enabledModels`               | Model patterns for cycling. Supports glob patterns (`github-copilot/*`, `*sonnet*`) and fuzzy matching. Same as `--models` CLI flag | -               |
 | `queueMode`                   | Message queue mode: `all` or `one-at-a-time`                                                                                        | `one-at-a-time` |

package/docs/rpc.md

@@ -23,6 +23,8 @@ Common options:
 - **Responses**: JSON objects with `type: "response"` indicating command success/failure
 - **Events**: Agent events streamed to stdout as JSON lines
 
+If you're consuming output in Bun, prefer `Bun.JSONL.parse(text)` for buffered JSONL or `Bun.JSONL.parseChunk()` for streaming output instead of splitting and `JSON.parse`.
+
 All commands support an optional `id` field for request/response correlation. If provided, the corresponding response will include the same `id`.
 
 ## Commands
@@ -1034,6 +1036,7 @@ Created by the `bash` RPC command (not by LLM tool calls):
 ```python
 import subprocess
 import json
+import jsonlines
 
 proc = subprocess.Popen(
     ["omp", "--mode", "rpc", "--no-session"],
@@ -1047,8 +1050,9 @@ def send(cmd):
     proc.stdin.flush()
 
 def read_events():
-    for line in proc.stdout:
-        yield json.loads(line)
+    with jsonlines.Reader(proc.stdout) as reader:
+        for event in reader:
+            yield event
 
 # Send prompt
 send({"type": "prompt", "message": "Hello!"})
@@ -1065,26 +1069,39 @@ for event in read_events():
         break
 ```
 
-## Example: Interactive Client (Node.js)
+## Example: Interactive Client (Bun)
 
 See [`test/rpc-example.ts`](../test/rpc-example.ts) for a complete interactive example, or [`src/modes/rpc/rpc-client.ts`](../src/modes/rpc/rpc-client.ts) for a typed client implementation.
 
 ```javascript
-const { spawn } = require("child_process");
-const readline = require("readline");
-
-const agent = spawn("omp", ["--mode", "rpc", "--no-session"]);
-
-readline.createInterface({ input: agent.stdout }).on("line", (line) => {
-	const event = JSON.parse(line);
+const agent = Bun.spawn(["omp", "--mode", "rpc", "--no-session"], {
+	stdin: "pipe",
+	stdout: "pipe",
+});
 
-	if (event.type === "message_update") {
-		const { assistantMessageEvent } = event;
-		if (assistantMessageEvent.type === "text_delta") {
-			process.stdout.write(assistantMessageEvent.delta);
+const decoder = new TextDecoder();
+let buffer = "";
+
+async function readEvents() {
+	const reader = agent.stdout.getReader();
+	while (true) {
+		const { value, done } = await reader.read();
+		if (done) break;
+		buffer += decoder.decode(value, { stream: true });
+		const result = Bun.JSONL.parseChunk(buffer);
+		buffer = buffer.slice(result.read);
+		for (const event of result.values) {
+			if (event.type === "message_update") {
+				const { assistantMessageEvent } = event;
+				if (assistantMessageEvent.type === "text_delta") {
+					process.stdout.write(assistantMessageEvent.delta);
+				}
+			}
 		}
 	}
-});
+}
+
+readEvents();
 
 // Send prompt
 agent.stdin.write(JSON.stringify({ type: "prompt", message: "Hello" }) + "\n");

package/docs/session.md

@@ -227,12 +227,10 @@ Entries form a tree:
 ## Parsing Example
 
 ```typescript
-import * as fs from "node:fs";
+const text = await Bun.file("session.jsonl").text();
+const entries = Bun.JSONL.parse(text);
 
-const lines = fs.readFileSync("session.jsonl", "utf8").trim().split("\n");
-
-for (const line of lines) {
-	const entry = JSON.parse(line);
+for (const entry of entries) {
 
 	switch (entry.type) {
 		case "session":

package/docs/tui.md

@@ -153,7 +153,7 @@ handleInput(data: string) {
 
 ## Line Width
 
-**Critical:** Each line from `render()` must not exceed the `width` parameter.
+**Critical:** Each line from `render()` must not exceed the `width` parameter. Width calculations and wrapping follow Bun’s built-ins (`Bun.stringWidth`, `Bun.wrapAnsi`).
 
 ```typescript
 import { visibleWidth, truncateToWidth } from "@oh-my-pi/pi-tui";
@@ -166,9 +166,9 @@ render(width: number): string[] {
 
 Utilities:
 
-- `visibleWidth(str)` - Get display width (ignores ANSI codes)
+- `visibleWidth(str)` - Get display width (ANSI-safe, Unicode-width aware)
 - `truncateToWidth(str, width, ellipsis?)` - Truncate with optional ellipsis
-- `wrapTextWithAnsi(str, width)` - Word wrap preserving ANSI codes
+- `wrapTextWithAnsi(str, width)` - Word wrap preserving ANSI codes (Bun.wrapAnsi)
 
 ## Creating Custom Components
 

package/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "@oh-my-pi/pi-coding-agent",
-	"version": "8.5.0",
+	"version": "8.8.8",
 	"description": "Coding agent CLI with read, bash, edit, write tools and session management",
 	"type": "module",
 	"ompConfig": {
@@ -83,11 +83,11 @@
 		"test": "bun test"
 	},
 	"dependencies": {
-		"@oh-my-pi/omp-stats": "8.5.0",
-		"@oh-my-pi/pi-agent-core": "8.5.0",
-		"@oh-my-pi/pi-ai": "8.5.0",
-		"@oh-my-pi/pi-tui": "8.5.0",
-		"@oh-my-pi/pi-utils": "8.5.0",
+		"@oh-my-pi/omp-stats": "8.8.8",
+		"@oh-my-pi/pi-agent-core": "8.8.8",
+		"@oh-my-pi/pi-ai": "8.8.8",
+		"@oh-my-pi/pi-tui": "8.8.8",
+		"@oh-my-pi/pi-utils": "8.8.8",
 		"@openai/agents": "^0.4.3",
 		"@sinclair/typebox": "^0.34.46",
 		"ajv": "^8.17.1",
@@ -100,7 +100,6 @@
 		"highlight.js": "^11.11.1",
 		"marked": "^17.0.1",
 		"nanoid": "^5.1.6",
-		"ndjson": "^2.0.0",
 		"node-html-parser": "^7.0.2",
 		"smol-toml": "^1.6.0",
 		"strip-ansi": "^7.1.2",
@@ -109,7 +108,6 @@
 	"devDependencies": {
 		"@types/diff": "^8.0.0",
 		"@types/ms": "^2.1.0",
-		"@types/ndjson": "^2.0.4",
 		"@types/node": "^25.0.10",
 		"ms": "^2.1.3"
 	},

package/src/cli/args.ts

@@ -15,6 +15,7 @@ export interface Args {
 	model?: string;
 	smol?: string;
 	slow?: string;
+	plan?: string;
 	apiKey?: string;
 	systemPrompt?: string;
 	appendSystemPrompt?: string;
@@ -85,6 +86,8 @@ export function parseArgs(args: string[], extensionFlags?: Map<string, { type: "
 			result.smol = args[++i];
 		} else if (arg === "--slow" && i + 1 < args.length) {
 			result.slow = args[++i];
+		} else if (arg === "--plan" && i + 1 < args.length) {
+			result.plan = args[++i];
 		} else if (arg === "--api-key" && i + 1 < args.length) {
 			result.apiKey = args[++i];
 		} else if (arg === "--system-prompt" && i + 1 < args.length) {
@@ -193,6 +196,7 @@ ${chalk.bold("Options:")}
   --model <pattern>              Model to use (fuzzy match: "opus", "gpt-5.2", or "p-openai/gpt-5.2")
    --smol <id>                    Smol/fast model for lightweight tasks (or OMP_SMOL_MODEL env)
    --slow <id>                    Slow/reasoning model for thorough analysis (or OMP_SLOW_MODEL env)
+   --plan <id>                    Plan model for architectural planning (or OMP_PLAN_MODEL env)
   --api-key <key>                API key (defaults to env vars)
   --system-prompt <text>         System prompt (default: coding assistant prompt)
   --append-system-prompt <text>  Append text or file contents to the system prompt

package/src/commit/agentic/prompts/analyze-file.md

@@ -19,4 +19,4 @@ Return a concise JSON object with:
 Consider how this file's changes relate to the above files.
 {{/if}}
 
-Call the complete tool with the JSON payload.
+Call the submit_result tool with the JSON payload.

package/src/commit/agentic/tools/analyze-file.ts

@@ -82,7 +82,7 @@ export function createAnalyzeFileTool(options: {
 			const taskParams: TaskParams = {
 				agent: "quick_task",
 				context,
-				output: analyzeFileOutputSchema,
+				schema: analyzeFileOutputSchema,
 				tasks,
 			};
 			return taskTool.execute(toolCallId, taskParams, signal, onUpdate);

package/src/config/prompt-templates.ts

@@ -1,4 +1,5 @@
 import * as path from "node:path";
+import { jtdToTypeScript } from "@oh-my-pi/pi-coding-agent/tools/jtd-to-typescript";
 import { logger } from "@oh-my-pi/pi-utils";
 import Handlebars from "handlebars";
 import { CONFIG_DIR_NAME, getPromptsDir } from "../config";
@@ -222,6 +223,8 @@ handlebars.registerHelper("includes", (collection: unknown, item: unknown): bool
  */
 handlebars.registerHelper("not", (value: unknown): boolean => !value);
 
+handlebars.registerHelper("jtdToTypeScript", (schema: unknown): string => jtdToTypeScript(schema));
+
 export function renderPromptTemplate(template: string, context: TemplateContext = {}): string {
 	const compiled = handlebars.compile(template, { noEscape: true, strict: false });
 	const rendered = compiled(context ?? {});

package/src/config/settings-manager.ts

@@ -13,6 +13,8 @@ export interface CompactionSettings {
 	enabled?: boolean; // default: true
 	reserveTokens?: number; // default: 16384
 	keepRecentTokens?: number; // default: 20000
+	autoContinue?: boolean; // default: true
+	remoteEndpoint?: string;
 }
 
 export interface BranchSummarySettings {
@@ -838,11 +840,27 @@ export class SettingsManager {
 		return this.settings.compaction?.keepRecentTokens ?? 20000;
 	}
 
-	getCompactionSettings(): { enabled: boolean; reserveTokens: number; keepRecentTokens: number } {
+	getCompactionAutoContinue(): boolean {
+		return this.settings.compaction?.autoContinue ?? true;
+	}
+
+	getCompactionRemoteEndpoint(): string | undefined {
+		return this.settings.compaction?.remoteEndpoint;
+	}
+
+	getCompactionSettings(): {
+		enabled: boolean;
+		reserveTokens: number;
+		keepRecentTokens: number;
+		autoContinue: boolean;
+		remoteEndpoint?: string;
+	} {
 		return {
 			enabled: this.getCompactionEnabled(),
 			reserveTokens: this.getCompactionReserveTokens(),
 			keepRecentTokens: this.getCompactionKeepRecentTokens(),
+			autoContinue: this.getCompactionAutoContinue(),
+			remoteEndpoint: this.getCompactionRemoteEndpoint(),
 		};
 	}
 

package/src/export/html/template.css

@@ -137,6 +137,15 @@
     }
 
     .tree-node.in-path {
+      background: color-mix(in srgb, var(--accent) 10%, transparent);
+    }
+
+    .tree-node:not(.in-path) {
+      opacity: 0.5;
+    }
+
+    .tree-node:not(.in-path):hover {
+      opacity: 1;
     }
 
     .tree-prefix {