byterover-cli 0.2.1 → 0.3.0

package/README.md

@@ -4,7 +4,7 @@ Command-line interface for ByteRover, enabling seamless team/space management, a
 
 [![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)
-[![Node](https://img.shields.io/badge/node-%3E%3D18.0.0-brightgreen.svg)](https://nodejs.org)
+[![Node](https://img.shields.io/badge/node-%3E%3D20.0.0-brightgreen.svg)](https://nodejs.org)
 
 ## Important
 
@@ -18,8 +18,7 @@ Please check the migration guide [here](https://docs.byterover.dev/beta/migratio
 
 * [Installation](#installation)
 * [Quick Start](#quick-start)
-* [What is ACE?](#what-is-ace)
-* [Core Workflow](#core-workflow)
+* [What is Context Tree?](#what-is-context-tree)
 * [Essential Commands](#essential-commands)
 * [Authentication](#authentication)
 * [Configuration](#configuration)
@@ -29,7 +28,7 @@ Please check the migration guide [here](https://docs.byterover.dev/beta/migratio
 
 ### Requirements
 
-- **Node.js**: >= 18.0.0
+- **Node.js**: >= 20.0.0
 - **Operating System**:
   - macOS
   - Windows
@@ -78,46 +77,26 @@ Select a space from your available spaces and configure your project.
 
 You're ready to use ByteRover commands in your project!
 
-## What is ACE?
+## What is Context Tree?
 
-**Agentic Context Engineering (ACE)** is a systematic workflow that helps coding agents (like Claude Code, Cursor, etc.) capture their work, learn from feedback, and build cumulative knowledge in a living playbook.
+The **Context Tree** is ByteRover's structured memory system that helps you and your coding agents organize, store, and retrieve project knowledge efficiently.
 
-Based on the research paper [**Agentic Context Engineering: Evolving Contexts for Self-Improving Language Models**](https://arxiv.org/abs/2510.04618) by Dang et al., ACE enables language models to iteratively improve their performance through structured context evolution.
+### Why Use Context Tree?
 
-### Why Use ACE?
+- **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 coding agents like Claude Code, Cursor, and others
+- **Version Control**: Push and sync your context to ByteRover's cloud storage
 
-- **Build Knowledge**: Each task completed by an agent teaches the system
-- **Learn from Experience**: Capture both successes and failures as insights
-- **Context Persistence**: Maintain project-specific best practices that improve over time
-- **Traceability**: Track what worked, what didn't, and why
+### How It Works
 
-## Core Workflow
+The context tree organizes knowledge into:
+- **Domains**: High-level categories (e.g., Architecture, API, Frontend)
+- **Topics**: Specific subjects within domains (e.g., Authentication, Components)
+- **Context Files**: Markdown files containing your actual knowledge
 
-ACE follows a simple 3-phase cycle that coding agents can use to improve over time:
-
-### 1. Executor
-Agent performs coding task and saves detailed output with context.
-
-### 2. Reflector
-Agent analyzes results and provides honest feedback on what worked and what didn't.
-
-### 3. Curator
-Agent transforms insights into playbook updates that are automatically applied to improve future work.
-
-### Quick Example
-
-If you're using a coding agent like Claude Code:
-
-```bash
-# Complete ACE workflow in a single command
-brv complete "auth-feature" \
-  "Implemented JWT authentication with secure token handling" \
-  "Successfully added OAuth2 authentication" \
-  --tool-usage "Read:src/auth.ts,Edit:src/auth.ts,Bash:npm test" \
-  --feedback "All tests passed, auth works correctly"
-```
-
-For comprehensive ACE instructions for coding agents, check the corresponding coding agents' instruction files after `brv init` or `brv gen-rules`.
+For comprehensive instructions for coding agents, check the generated rule files after running `brv gen-rules`.
 
 ## Essential Commands
 
@@ -127,6 +106,9 @@ For comprehensive ACE instructions for coding agents, check the corresponding co
 # Log in to ByteRover
 brv login
 
+# Log out and clear credentials
+brv logout
+
 # Check your authentication and project status
 brv status
 ```
@@ -147,23 +129,27 @@ brv space switch
 ### Memory Operations
 
 ```bash
-# Retrieve memories from ByteRover (outputs to stdout for agent context)
-brv retrieve --query "authentication best practices"
-brv retrieve -q "error handling" -n "src/auth/login.ts,src/auth/oauth.ts"
+# Add content to context tree (interactive or autonomous)
+brv curate
+brv curate "User authentication uses JWT tokens"
 
-# Push your playbook to ByteRover's memory storage
+# Push your context tree to ByteRover's memory storage
 brv push
+
+# Pull context tree from ByteRover cloud
+brv pull
+
+# Query the context tree
+brv query "How is authentication implemented?"
+
+# Reset context tree to default state
+brv clear
 ```
 
 ### For Coding Agents
 
 ```bash
-# Complete ACE workflow (recommended for agents)
-brv complete <hint> <reasoning> <answer> \
-  --tool-usage <tools> \
-  --feedback <feedback>
-
-# Generate agent rules (sets up ACE workflow for your coding agent)
+# Generate agent rules (sets up context tree workflow for your coding agent)
 brv gen-rules
 ```
 
@@ -195,19 +181,30 @@ When you run `brv init`, a configuration file is created at `.brv/config.json` i
 - **User information**: Your user ID and email
 - **Project settings**: Project-specific configuration
 
-### ACE File Structure
+### Context Tree Structure
 
-ACE stores all outputs in `.brv/ace/`:
+The context tree is stored in `.brv/context-tree/`:
 
 ```
-.brv/ace/
-├── playbook.json                # Your living knowledge base
-├── executor-outputs/            # Coding task outputs
-├── reflections/                 # Task analysis and feedback
-└── deltas/                      # Playbook updates
+.brv/context-tree/
+├── Architecture/
+│   ├── System Design/
+│   │   └── context.md
+│   └── Database Schema/
+│       └── context.md
+├── API/
+│   ├── Authentication/
+│   │   └── context.md
+│   └── Endpoints/
+│       └── context.md
+└── Frontend/
+    ├── Components/
+    │   └── context.md
+    └── State Management/
+        └── context.md
 ```
 
-**Note**: When you run `brv push`, the playbook is uploaded to ByteRover's memory storage, and local ACE files are automatically cleaned up to keep your workspace organized.
+**Note**: When you run `brv push`, your context tree is uploaded to ByteRover's memory storage for version control and team collaboration.
 
 ## Getting Help
 
@@ -221,6 +218,9 @@ brv --help
 brv login --help
 brv init --help
 brv push --help
+brv pull --help
+brv query --help
+brv clear --help
 ```
 
 ### Support

package/bin/dev.js

@@ -1,4 +1,4 @@
-#!/usr/bin/env -S node --loader ts-node/esm --disable-warning=ExperimentalWarning
+#!/usr/bin/env -S node --loader ts-node/esm --no-warnings=ExperimentalWarning
 
 process.env.BRV_ENV = 'development'
 

package/dist/commands/cipher-agent/run.d.ts

@@ -0,0 +1,111 @@
+import { Command } from '@oclif/core';
+import type { IProjectConfigStore } from '../../core/interfaces/i-project-config-store.js';
+export default class CipherAgentRun extends Command {
+    static args: {
+        prompt: import("@oclif/core/interfaces").Arg<string | undefined, Record<string, unknown>>;
+    };
+    static description: string;
+    static examples: string[];
+    static flags: {
+        continue: import("@oclif/core/interfaces").BooleanFlag<boolean>;
+        interactive: import("@oclif/core/interfaces").BooleanFlag<boolean>;
+        maxTokens: import("@oclif/core/interfaces").OptionFlag<number | undefined, import("@oclif/core/interfaces").CustomOptions>;
+        resume: import("@oclif/core/interfaces").OptionFlag<string | undefined, import("@oclif/core/interfaces").CustomOptions>;
+        temperature: import("@oclif/core/interfaces").OptionFlag<string | undefined, import("@oclif/core/interfaces").CustomOptions>;
+        workingDirectory: import("@oclif/core/interfaces").OptionFlag<string | undefined, import("@oclif/core/interfaces").CustomOptions>;
+        apiKey?: import("@oclif/core/interfaces").OptionFlag<string | undefined, import("@oclif/core/interfaces").CustomOptions> | undefined;
+        model?: import("@oclif/core/interfaces").OptionFlag<string | undefined, import("@oclif/core/interfaces").CustomOptions> | undefined;
+        verbose?: import("@oclif/core/interfaces").BooleanFlag<boolean> | undefined;
+    };
+    static hidden: boolean;
+    catch(error: Error & {
+        oclif?: {
+            exit: number;
+        };
+    }): Promise<void>;
+    protected createServices(): {
+        projectConfigStore: IProjectConfigStore;
+    };
+    /**
+     * Generate a unique session ID with timestamp and random component.
+     *
+     * Format: "session-{timestamp}-{random}"
+     * Example: "session-1731686400123-a7b3c9"
+     *
+     * @returns Unique session ID string
+     */
+    protected generateSessionId(): string;
+    /**
+     * Get the most recent session ID from history storage.
+     *
+     * @param agent - CipherAgent instance
+     * @returns Most recent session ID or undefined if no sessions found
+     */
+    protected getMostRecentSessionId(agent: import('../../infra/cipher/cipher-agent.js').CipherAgent): Promise<string | undefined>;
+    run(): Promise<void>;
+    /**
+     * Validate that a session exists in storage.
+     *
+     * @param agent - CipherAgent instance
+     * @param sessionId - Session ID to validate
+     * @returns True if session exists
+     */
+    protected validateSessionExists(agent: import('../../infra/cipher/cipher-agent.js').CipherAgent, sessionId: string): Promise<boolean>;
+    /**
+     * Create LLM configuration from flags and environment
+     *
+     * @param token - Authentication token
+     * @param token.accessToken - Access token for authentication
+     * @param token.sessionKey - Session key for authentication
+     * @param token.spaceId - Space ID for the session
+     * @param token.teamId - Team ID for the session
+     * @param flags - Command flags
+     * @param flags.apiKey - OpenRouter API key for direct service (optional)
+     * @param flags.maxTokens - Maximum tokens in response
+     * @param flags.model - Model to use
+     * @param flags.temperature - Temperature for randomness
+     * @param flags.verbose - Enable verbose debug output
+     * @param flags.workingDirectory - Working directory for file operations
+     * @returns LLM configuration object
+     */
+    private createLLMConfig;
+    /**
+     * Format tool call for concise display in interactive mode
+     *
+     * @param toolName - Name of the tool
+     * @param args - Tool arguments
+     * @returns Formatted string for display
+     */
+    private formatToolForInteractive;
+    /**
+     * Format tool result summary for display
+     *
+     * @param toolName - Name of the tool
+     * @param result - Tool result data
+     * @returns Formatted summary string or empty if no summary needed
+     */
+    private formatToolResultSummary;
+    /**
+     * Handle workspace not initialized error with friendly message
+     *
+     * @param _error - WorkspaceNotInitializedError instance (unused, kept for type safety)
+     */
+    private handleWorkspaceError;
+    /**
+     * Resolve session ID based on flags
+     *
+     * @param agent - CipherAgent instance
+     * @param flags - Command flags
+     * @param flags.continue - Continue most recent session flag
+     * @param flags.resume - Resume specific session flag
+     * @returns Resolved session ID
+     */
+    private resolveSessionId;
+    /**
+     * Setup event listeners based on mode
+     *
+     * @param agent - CipherAgent instance
+     * @param isInteractive - Whether in interactive mode
+     */
+    private setupEventListeners;
+}