AI Assistant Introduction
About 421 wordsAbout 1 min
AI Assistant Introduction collects the AI capabilities built into the CNB platform.
When to use: when you want AI help in your workflow but don't know where to start — for example letting AI take over an issue, analyzing build failures, asking questions against a repository knowledge base, or generating Wiki docs on demand. The CNB platform has several built-in AI features that boost developer productivity.
Features
CodeBuddy NPC
CodeBuddy NPC is the AI colleague living in your CNB dev infrastructure. You set the goal and review its output; it is a Cloud Agent built on CodeBuddy, deeply integrated with the CNB platform, acquiring context on its own and completing dev tasks end-to-end.
Core Capabilities:
- Goal-Driven: You define "what", the NPC handles "how". From requirements to PR, fully autonomous.
- Autonomous Context: Deeply integrated with CNB repos, issues, and pipelines. Understands requirements, reads code, checks logs on its own.
- Parallel Assignment: Dispatch several tasks at once. Multiple NPCs work in parallel in the cloud, unblocked and independently delivered.
Quick Start:
- Mention
@CodeBuddyin an Issue or PR comment, and the NPC automatically executes the task and replies - Define a role in your repository's
.cnb/settings.ymlto create your own NPC, then use it via@repo-path(role-name) - After enabling Work Mode, the NPC can autonomously write code, push code, and create PRs
For more information on event triggers and custom NPCs, see NPC.
TencentCloud CodeBuddy
TencentCloud CodeBuddy is built into the Workspaces WebIDE, ready to use without additional installation.
AI Chat
The platform has a built-in AI Chat, supporting quick access to help documentation and configuration file writing through AI.
RAG Knowledge Base
Set up enterprise or personal knowledge bases through code repositories: commit documents to the repository and configure the pipeline, and the content is processed by the large model and written to the knowledge base automatically, for webpage Q&A, OpenAPI, and other scenarios — helping build RAG (Retrieval-Augmented Generation) applications. See Knowledge Base for details.
Issue Analysis
Submit build error messages to AI with one click to identify the cause quickly.
Change Analysis
Submit commit or pull request change information to AI with one click to understand the changes and their potential impact quickly.