byterover-cli 1.2.0 → 1.3.0

package/README.md

@@ -1,6 +1,6 @@
 # ByteRover CLI
 
-Command-line interface for ByteRover, featuring an interactive REPL with a modern React/Ink terminal UI for managing your project's context tree and knowledge storage. Seamlessly integrate with 18+ AI coding agents including Claude Code, Cursor, Windsurf, GitHub Copilot, Cline, and more.
+Command-line interface for ByteRover, featuring an interactive REPL with a modern React/Ink terminal UI for managing your project's context tree and knowledge storage. Seamlessly integrate with 18 AI coding agents via modern skill files, MCP tools, or rules-based integration—supports Claude Code, Cursor, Windsurf, GitHub Copilot, Cline, and 13 more.
 
 [![Version](https://img.shields.io/npm/v/byterover-cli.svg)](https://npmjs.org/package/byterover-cli)
 [![Downloads/week](https://img.shields.io/npm/dw/byterover-cli.svg)](https://npmjs.org/package/byterover-cli)
@@ -11,7 +11,9 @@ Command-line interface for ByteRover, featuring an interactive REPL with a moder
 * [Installation](#installation)
 * [Quick Start](#quick-start)
 * [Interactive REPL](#interactive-repl)
+* [Keyboard Shortcuts](#keyboard-shortcuts)
 * [What is Context Tree?](#what-is-context-tree)
+* [Supported AI Agents](#supported-ai-agents)
 * [Slash Commands Reference](#slash-commands-reference)
 * [Authentication](#authentication)
 * [Configuration](#configuration)
@@ -78,6 +80,17 @@ This opens your browser to complete OAuth authentication. Your credentials are s
 
 Select a team and space from your available options, and ByteRover will set up your project's context tree.
 
+### 4. (Optional) Configure Agent Connectors
+
+```
+/connectors
+```
+
+ByteRover automatically configures the best connector for your installed agents:
+- **Skill files** for Claude Code and Cursor (modern, discoverable)
+- **MCP tools** for most other agents (universal protocol)
+- Switch connector types anytime via `/connectors`
+
 You're now ready to use ByteRover! Try `/status` to see your project's current state.
 
 ## Interactive REPL
@@ -99,10 +112,26 @@ The terminal UI includes:
 - **Tab Navigation**: Switch between Chat and Activity views using `Tab`
 - **Command Completion**: Type `/` to see available commands with auto-completion
 - **Activity Log**: Real-time task status and execution progress
-- **Streaming Output**: Live responses from AI-powered operations
-- **File References**: Type `@` in curate mode to browse and attach files (supports PDF)
+- **Streaming Output**: Live responses with markdown rendering (headings, lists, blockquotes, code blocks)
+- **Reasoning Display**: View agent thinking process with streamed reasoning blocks
+- **File References**: Type `@` in curate mode to browse and attach files
+- **PDF Support**: Reference and extract text from PDF files using `@` (up to 100 pages)
 - **Dynamic Domains**: Automatically creates new knowledge domains as your context tree grows
 - **Session Persistence**: Sessions auto-resume after restart
+- **Expandable Views**: Press `Ctrl+O` to expand messages or logs to full-screen with vim-style navigation
+- **Version Indicator**: Shows "(latest)" when running the most current version
+
+### Keyboard Shortcuts
+
+| Shortcut | Action |
+|----------|--------|
+| `Tab` | Switch between Chat and Activity views |
+| `Ctrl+O` | Expand message or log to full-screen |
+| `j` / `k` | Scroll down/up in expanded view |
+| `g` / `G` | Jump to top/bottom in expanded view |
+| `Esc` or `q` | Exit expanded view |
+| `/` | Show command suggestions |
+| `@` | Browse files (in curate mode) |
 
 ### Using Commands
 
@@ -125,7 +154,7 @@ The **Context Tree** is ByteRover's structured knowledge system that helps you a
 - **Organized Knowledge**: Structure your project knowledge by domain and topic
 - **Easy Retrieval**: Find relevant context quickly when you need it
 - **Persistent Memory**: Maintain project-specific knowledge across sessions
-- **Agent-Friendly**: Works seamlessly with 18+ AI coding agents (Claude Code, Cursor, Windsurf, GitHub Copilot, Cline, and more) via rules-based or hook-based integration
+- **Agent-Friendly**: Works seamlessly with 18 AI coding agents (Claude Code, Cursor, Windsurf, GitHub Copilot, Cline, and 13 more) via skill files, MCP tools, hooks, or rules
 - **Cloud Sync**: Push and sync your context tree to ByteRover's cloud storage for backup and team collaboration
 - **Dynamic Domains**: Automatically creates new domains as your knowledge grows
 
@@ -136,6 +165,20 @@ The context tree organizes knowledge into:
 - **Topics**: Specific subjects within domains (e.g., Authentication, Components)
 - **Context Files**: Markdown files containing your actual knowledge
 
+## Supported AI Agents
+
+ByteRover integrates with 18 AI coding agents:
+
+**Skill Connector (Default):**
+- Claude Code, Cursor
+
+**MCP Connector (Default):**
+- Amp, Augment Code, Cline, Gemini CLI, Github Copilot, Junie, Kilo Code, Kiro, Qoder, Qwen Code, Roo Code, Trae.ai, Warp, Windsurf, Zed (and Codex via global scope)
+
+**All agents support rules-based integration as a universal fallback option.**
+
+Use `/connectors` in the REPL to view, install, or switch connector types for your agents.
+
 ### Integrating with Coding Agents
 
 Use `/connectors` to manage integrations with your AI coding agents:
@@ -144,10 +187,17 @@ Use `/connectors` to manage integrations with your AI coding agents:
 /connectors
 ```
 
-ByteRover supports three connector types:
-- **Hook integration** (Claude Code): Direct injection via IDE settings for seamless integration
-- **Rules-based** (all agents): Generates agent-specific rule files (e.g., `CLAUDE.md`, `.cursorrules`) that instruct the agent how to read from and contribute to your context tree
-- **MCP integration** (Model Context Protocol): Exposes `brv-query` and `brv-curate` as MCP tools that AI agents can call directly
+ByteRover supports four connector types:
+
+1. **Skill integration** (Claude Code, Cursor - default): Modern integration that writes 3 markdown files (SKILL.md, TROUBLESHOOTING.md, WORKFLOWS.md) to your agent's skills directory for easy discovery and guidance
+2. **MCP integration** (16 other agents - default): Exposes brv-query and brv-curate as Model Context Protocol tools that AI agents can call directly
+3. **Rules-based** (all agents): Generates agent-specific rule files (e.g., CLAUDE.md, .cursorrules) with instructions for using ByteRover
+4. **Hook integration** (Claude Code only - legacy): Direct injection via IDE settings, replaced by skill connector
+
+**Defaults by agent:**
+- Claude Code, Cursor: Skill connector
+- All others (16 agents): MCP connector
+- Rules: Available for all agents as fallback
 
 ## Slash Commands Reference
 
@@ -208,9 +258,19 @@ ByteRover supports three connector types:
 
 | Command | Description |
 |---------|-------------|
-| `/connectors` | Manage agent connectors (rules-based or hook integration) |
+| `/connectors` | Manage agent connectors (skill, hook, mcp, or rules integration) |
 | `/reset [-y] [directory]` | Reset context tree to empty state |
 
+**Connector types:**
+- `skill`: Modern integration with markdown skill files (SKILL.md, TROUBLESHOOTING.md, WORKFLOWS.md)
+- `mcp`: Model Context Protocol tools (brv-query, brv-curate)
+- `hook`: Legacy IDE settings injection (Claude Code only)
+- `rules`: Agent-specific rule files (universal fallback)
+
+**Defaults:**
+- Claude Code, Cursor: `skill`
+- All others: `mcp`
+
 **Reset options:**
 - `-y, --yes`: Skip confirmation prompt
 
@@ -304,6 +364,24 @@ The context tree is stored in `.brv/context-tree/`:
 
 **Note**: When you run `/push`, your context tree is uploaded to ByteRover's cloud storage for version control and team collaboration.
 
+### Agent Connector Configuration
+
+Connector configurations are stored based on type:
+
+**Skill connectors** (Claude Code, Cursor):
+- **Project-scoped**: `.claude/skills/byterover/`, `.cursor/skills/byterover/`
+- **Global-scoped** (Codex): `~/.codex/skills/byterover/`
+- Files: `SKILL.md`, `TROUBLESHOOTING.md`, `WORKFLOWS.md`
+
+**Hook connectors** (Claude Code legacy):
+- `.claude/settings.local.json` (project-scoped)
+
+**MCP connectors**:
+- Managed via MCP configuration files (JSON/TOML)
+
+**Rules connectors**:
+- `CLAUDE.md`, `.cursorrules`, `.windsurfrules`, etc. (varies by agent)
+
 ## Troubleshooting
 
 ### Session Logs

package/dist/constants.d.ts

@@ -19,11 +19,6 @@ export declare const SNAPSHOT_FILE = ".snapshot.json";
  * This is ByteRover's internal branching mechanism, not Git branches.
  */
 export declare const DEFAULT_BRANCH = "main";
-/**
- * ByteRover documentation URL.
- * Used in CLI help output to direct users to online documentation.
- */
-export declare const DOCS_URL = "https://docs.byterover.dev";
 export declare const TRANSPORT_HOST = "127.0.0.1";
 export declare const TRANSPORT_REQUEST_TIMEOUT_MS = 10000;
 export declare const TRANSPORT_ROOM_TIMEOUT_MS = 2000;

package/dist/constants.js

@@ -24,11 +24,6 @@ export const SNAPSHOT_FILE = '.snapshot.json';
  * This is ByteRover's internal branching mechanism, not Git branches.
  */
 export const DEFAULT_BRANCH = 'main';
-/**
- * ByteRover documentation URL.
- * Used in CLI help output to direct users to online documentation.
- */
-export const DOCS_URL = 'https://docs.byterover.dev';
 // Transport layer constants (optimized for localhost real-time)
 export const TRANSPORT_HOST = '127.0.0.1'; // Use hostname for better sandbox compatibility
 export const TRANSPORT_REQUEST_TIMEOUT_MS = 10_000; // 10s - most operations complete quickly

package/dist/core/domain/cipher/agent/agent-info.d.ts

@@ -33,11 +33,11 @@ export declare const AgentPermissionSchema: z.ZodObject<{
      */
     edit: z.ZodDefault<z.ZodEnum<["allow", "deny", "ask"]>>;
 }, "strip", z.ZodTypeAny, {
-    edit: "allow" | "deny" | "ask";
     bash: Record<string, "allow" | "deny" | "ask">;
+    edit: "allow" | "deny" | "ask";
 }, {
-    edit?: "allow" | "deny" | "ask" | undefined;
     bash?: Record<string, "allow" | "deny" | "ask"> | undefined;
+    edit?: "allow" | "deny" | "ask" | undefined;
 }>;
 export type AgentPermission = z.infer<typeof AgentPermissionSchema>;
 /**
@@ -115,11 +115,11 @@ export declare const AgentInfoSchema: z.ZodObject<{
          */
         edit: z.ZodDefault<z.ZodEnum<["allow", "deny", "ask"]>>;
     }, "strip", z.ZodTypeAny, {
-        edit: "allow" | "deny" | "ask";
         bash: Record<string, "allow" | "deny" | "ask">;
+        edit: "allow" | "deny" | "ask";
     }, {
-        edit?: "allow" | "deny" | "ask" | undefined;
         bash?: Record<string, "allow" | "deny" | "ask"> | undefined;
+        edit?: "allow" | "deny" | "ask" | undefined;
     }>>;
     /**
      * Inline system prompt for this agent.
@@ -145,44 +145,44 @@ export declare const AgentInfoSchema: z.ZodObject<{
      */
     tools: z.ZodDefault<z.ZodRecord<z.ZodString, z.ZodBoolean>>;
 }, "strip", z.ZodTypeAny, {
-    name: string;
     hidden: boolean;
     mode: "primary" | "subagent" | "all";
+    name: string;
     native: boolean;
     permission: {
-        edit: "allow" | "deny" | "ask";
         bash: Record<string, "allow" | "deny" | "ask">;
+        edit: "allow" | "deny" | "ask";
     };
     tools: Record<string, boolean>;
-    prompt?: string | undefined;
+    color?: string | undefined;
+    description?: string | undefined;
     maxIterations?: number | undefined;
     model?: {
         modelId: string;
         providerId?: string | undefined;
     } | undefined;
-    description?: string | undefined;
-    color?: string | undefined;
-    temperature?: number | undefined;
+    prompt?: string | undefined;
     promptFile?: string | undefined;
+    temperature?: number | undefined;
 }, {
-    name: string;
     mode: "primary" | "subagent" | "all";
-    prompt?: string | undefined;
+    name: string;
+    color?: string | undefined;
+    description?: string | undefined;
+    hidden?: boolean | undefined;
     maxIterations?: number | undefined;
     model?: {
         modelId: string;
         providerId?: string | undefined;
     } | undefined;
-    description?: string | undefined;
-    hidden?: boolean | undefined;
-    color?: string | undefined;
-    temperature?: number | undefined;
     native?: boolean | undefined;
     permission?: {
-        edit?: "allow" | "deny" | "ask" | undefined;
         bash?: Record<string, "allow" | "deny" | "ask"> | undefined;
+        edit?: "allow" | "deny" | "ask" | undefined;
     } | undefined;
+    prompt?: string | undefined;
     promptFile?: string | undefined;
+    temperature?: number | undefined;
     tools?: Record<string, boolean> | undefined;
 }>;
 /**

package/dist/core/domain/cipher/llm/schemas.d.ts

@@ -19,16 +19,16 @@ export declare const LLMConfigBaseSchema: z.ZodObject<{
     maxIterations: number;
     model: string;
     provider: "claude" | "gemini" | "openrouter";
-    timeout?: number | undefined;
     temperature?: number | undefined;
+    timeout?: number | undefined;
     maxInputTokens?: number | undefined;
     maxOutputTokens?: number | undefined;
 }, {
     model: string;
     provider: "claude" | "gemini" | "openrouter";
-    timeout?: number | undefined;
     maxIterations?: number | undefined;
     temperature?: number | undefined;
+    timeout?: number | undefined;
     maxInputTokens?: number | undefined;
     maxOutputTokens?: number | undefined;
 }>;
@@ -47,32 +47,32 @@ export declare const LLMConfigSchema: z.ZodEffects<z.ZodObject<{
     maxIterations: number;
     model: string;
     provider: "claude" | "gemini" | "openrouter";
-    timeout?: number | undefined;
     temperature?: number | undefined;
+    timeout?: number | undefined;
     maxInputTokens?: number | undefined;
     maxOutputTokens?: number | undefined;
 }, {
     model: string;
     provider: "claude" | "gemini" | "openrouter";
-    timeout?: number | undefined;
     maxIterations?: number | undefined;
     temperature?: number | undefined;
+    timeout?: number | undefined;
     maxInputTokens?: number | undefined;
     maxOutputTokens?: number | undefined;
 }>, {
     maxIterations: number;
     model: string;
     provider: "claude" | "gemini" | "openrouter";
-    timeout?: number | undefined;
     temperature?: number | undefined;
+    timeout?: number | undefined;
     maxInputTokens?: number | undefined;
     maxOutputTokens?: number | undefined;
 }, {
     model: string;
     provider: "claude" | "gemini" | "openrouter";
-    timeout?: number | undefined;
     maxIterations?: number | undefined;
     temperature?: number | undefined;
+    timeout?: number | undefined;
     maxInputTokens?: number | undefined;
     maxOutputTokens?: number | undefined;
 }>;
@@ -97,37 +97,37 @@ export declare const LLMUpdatesSchema: z.ZodEffects<z.ZodObject<{
     temperature: z.ZodOptional<z.ZodOptional<z.ZodNumber>>;
     timeout: z.ZodOptional<z.ZodOptional<z.ZodNumber>>;
 }, "strip", z.ZodTypeAny, {
-    timeout?: number | undefined;
     maxIterations?: number | undefined;
     model?: string | undefined;
-    provider?: "claude" | "gemini" | "openrouter" | undefined;
     temperature?: number | undefined;
+    timeout?: number | undefined;
     maxInputTokens?: number | undefined;
     maxOutputTokens?: number | undefined;
+    provider?: "claude" | "gemini" | "openrouter" | undefined;
 }, {
-    timeout?: number | undefined;
     maxIterations?: number | undefined;
     model?: string | undefined;
-    provider?: "claude" | "gemini" | "openrouter" | undefined;
     temperature?: number | undefined;
+    timeout?: number | undefined;
     maxInputTokens?: number | undefined;
     maxOutputTokens?: number | undefined;
+    provider?: "claude" | "gemini" | "openrouter" | undefined;
 }>, {
-    timeout?: number | undefined;
     maxIterations?: number | undefined;
     model?: string | undefined;
-    provider?: "claude" | "gemini" | "openrouter" | undefined;
     temperature?: number | undefined;
+    timeout?: number | undefined;
     maxInputTokens?: number | undefined;
     maxOutputTokens?: number | undefined;
+    provider?: "claude" | "gemini" | "openrouter" | undefined;
 }, {
-    timeout?: number | undefined;
     maxIterations?: number | undefined;
     model?: string | undefined;
-    provider?: "claude" | "gemini" | "openrouter" | undefined;
     temperature?: number | undefined;
+    timeout?: number | undefined;
     maxInputTokens?: number | undefined;
     maxOutputTokens?: number | undefined;
+    provider?: "claude" | "gemini" | "openrouter" | undefined;
 }>;
 /**
  * Input type for LLM updates.

package/dist/core/domain/cipher/session/session-metadata.d.ts

@@ -88,13 +88,13 @@ export declare const ActiveSessionPointerSchema: z.ZodObject<{
     sessionId: z.ZodString;
 }, "strip", z.ZodTypeAny, {
     sessionId: string;
-    pid: number;
     activatedAt: string;
+    pid: number;
     processToken?: string | undefined;
 }, {
     sessionId: string;
-    pid: number;
     activatedAt: string;
+    pid: number;
     processToken?: string | undefined;
 }>;
 /**

package/dist/core/domain/entities/agent.js

@@ -35,8 +35,8 @@ export const AGENT_CONNECTOR_CONFIG = {
         supported: ['rules', 'mcp'],
     },
     'Claude Code': {
-        default: 'hook',
-        supported: ['rules', 'hook', 'mcp'],
+        default: 'skill',
+        supported: ['rules', 'hook', 'mcp', 'skill'],
     },
     Cline: {
         default: 'mcp',
@@ -44,11 +44,11 @@ export const AGENT_CONNECTOR_CONFIG = {
     },
     Codex: {
         default: 'mcp',
-        supported: ['rules', 'mcp'],
+        supported: ['rules', 'mcp', 'skill'],
     },
     Cursor: {
-        default: 'mcp',
-        supported: ['rules', 'mcp'],
+        default: 'skill',
+        supported: ['rules', 'mcp', 'skill'],
     },
     'Gemini CLI': {
         default: 'mcp',
@@ -56,7 +56,7 @@ export const AGENT_CONNECTOR_CONFIG = {
     },
     'Github Copilot': {
         default: 'mcp',
-        supported: ['rules', 'mcp'],
+        supported: ['rules', 'mcp', 'skill'],
     },
     Junie: {
         default: 'mcp',

package/dist/core/domain/entities/connector-type.d.ts

@@ -5,6 +5,7 @@
  * - 'rules': Agent reads instructions from a rule file (e.g., CLAUDE.md)
  * - 'hook': Instructions are injected on each prompt via agent hooks
  * - 'mcp': Agent uses mcp tools to interact with brv
+ * - 'skill': Agent reads skill files from a project subdirectory
  */
-export declare const CONNECTOR_TYPES: readonly ["rules", "hook", "mcp"];
+export declare const CONNECTOR_TYPES: readonly ["rules", "hook", "mcp", "skill"];
 export type ConnectorType = (typeof CONNECTOR_TYPES)[number];

package/dist/core/domain/entities/connector-type.js

@@ -5,5 +5,6 @@
  * - 'rules': Agent reads instructions from a rule file (e.g., CLAUDE.md)
  * - 'hook': Instructions are injected on each prompt via agent hooks
  * - 'mcp': Agent uses mcp tools to interact with brv
+ * - 'skill': Agent reads skill files from a project subdirectory
  */
-export const CONNECTOR_TYPES = ['rules', 'hook', 'mcp'];
+export const CONNECTOR_TYPES = ['rules', 'hook', 'mcp', 'skill'];