proma-ai/Proma — 把最丝滑的通用 Agent 体验带进你的工作流，为 100x 专业用户而生的未来产品，正在实现 proactive Agent 阶段。基于 Claude Agent SDK 的完整开源实践，原生支持飞…

Description and use

**Brief Summary**
Proma is an open‑source, Claude‑Agent‑SDK‑powered framework that brings a highly responsive, proactive AI agent into everyday workflows. It natively supports Feishu group chat invocation and can plug into any large‑model provider, letting power users automate repetitive tasks, orchestrate toolchains, and schedule operations from the apps they already use.

**Value**

• **Automation of manual work** – Proma’s agent can act as a “smart assistant” that watches a workflow, triggers actions, and replies in real time, eliminating the need for humans to copy‑paste data, click through UI steps, or manually schedule jobs.
• **Vendor‑agnostic model layer** – By abstracting the underlying LLM behind a unified SDK, teams can swap Claude, GPT‑4, or any other provider without rewriting business logic, protecting future AI‑budget decisions.
• **Native collaboration integration** – Direct Feishu group‑chat hooks let the agent surface insights and take commands where the team already communicates, reducing context‑switching and increasing adoption.

**Practical Adoption Path**

1. **Prototype** – Clone the repo, run the provided Docker/CLI starter, and connect a Claude (or alternative) API key via the SDK config.
2. **Integrate** – Use the TypeScript SDK to wrap existing scripts or APIs (e.g., ticketing, CI/CD, CRM) as “agent actions”.
3. **Deploy** – Deploy the agent as a lightweight service (Docker, serverless, or within your internal Kubernetes cluster) and register the Feishu webhook URL to enable group‑chat interaction.
4. **Iterate** – Leverage the SDK’s “proactive” callbacks to schedule recurring jobs or react to external events, then gradually expand the set of automated flows.

**Production Readiness**

• **Activity & Community** – 1,322 stars, 180 forks, recent commits (as of 2026‑06‑23) and active issue discussion indicate a healthy, maintained project.
• **Technical Maturity** – Written in TypeScript with a clear SDK, CLI, and API surface; supports multiple LLM providers and has built‑in Feishu integration, reducing custom glue code.
• **Scalability** – The service can be containerized and horizontally scaled; the proactive agent model is designed for low‑latency, event‑driven use cases.
• **Risk Considerations** – License compliance, security hardening, and long‑term maintainer commitment still need a final review, but no major red flags appear.

Overall, Proma is a production‑ready OSS candidate for teams looking to embed proactive AI agents into their daily tooling, with a straightforward path from prototype to full‑scale deployment.

Problem

proma-ai/Proma helps remove repetitive manual operations from a workflow.

Outcome

High for an OSS candidate: recent activity, adoption, and ecosystem signals are strong enough for a serious pilot.

Workflow

• remove manual work
• connect tools into repeatable flows
• schedule operational tasks

Evidence

把最丝滑的通用 Agent 体验带进你的工作流，为 100x 专业用户而生的未来产品，正在实现 proactive Agent 阶段。基于 Claude Agent SDK 的完整开源实践，原生支持飞书群聊调用、灵活接入任意大模型供应商 —— 让顶级 Agent 能力真正跑在你每天用的地方。

Source