How to bring your Claude Code or Cursor agent into team chat

说明与用途

代码代理现在可以通过 Pachca CLI 和令牌获取聊天上下文、发布消息并自动执行任务，消除了手动构造提示的需要。

问题

Coding agents need contextual information from team chat (bug reports, decisions, checklists) that is not in the repository.

结果

Agents can now access real‑time chat data, post updates, and run automated workflows without manual context injection, improving efficiency and reducing errors.

工作流

• Install Pachca CLI (Node.js 20+).
• Sign in with pachca auth login to obtain a short‑lived token stored in the OS keychain.
• Use CLI commands like pachca api ls, pachca api POST /messages --describe, pachca api POST /messages --docs to explore the API.
• Ask the agent to perform tasks such as reading recent messages, summarizing threads, or posting release notes using commands like pachca messages create.
• Optionally add Agent Skills with npx skills add pachca/openapi to pre‑configure common workflows.
• For CI or headless agents, create a bot token and use PACHCA_TOKEN=$PACHCA_SECRET pachca messages create --no-input to avoid interactive login.

证据

Coding agents have become good at reading code, running commands and opening pull requests. But most of the context they need doesn't live in the repository.

来源

The Pachca CLI needs Node.js 20 or later. You can run it without installing, which works well for agents with temporary environments:

来源

On your own machine, sign in through the browser: pachca auth login

来源

Now ask your agent to do something real. It doesn't need instructions on the API, because it can look things up itself:

来源

For CI and headless mode, create a token in Pachca in advance and pass it through an environment variable from your secrets:

来源