buildswithpaul/Frappe_Assistant_Core — Infrastructure that connects LLMs to ERPNext.

Description and use

**Brief Summary (2‑3 sentences)**
Frappe Assistant Core is an open‑source infrastructure layer that links large language models (LLMs) to ERPNext via the Model Context Protocol (MCP). By exposing ERPNext’s APIs through a standardized MCP server, it lets any compatible LLM invoke real‑world business functions—such as creating invoices, fetching inventory, or running workflows—without custom glue code.

**Value**

• **Standardized integration**: MCP provides a vendor‑agnostic contract, so the same assistant code can work with ERPNext, other Frappe apps, or future tools that adopt the protocol.
• **Rapid AI‑enabled automation**: Developers can turn ERPNext features into “actions” that an LLM can call, enabling conversational assistants, autonomous agents, or workflow bots with minimal boilerplate.
• **Reusable ecosystem**: The core includes ready‑made API/SDK/CLI hooks, language metadata, and topic definitions, lowering the effort to ship new AI‑driven services on top of ERPNext.

**Practical Adoption Path**

1. **Clone & install** the repository (Python‑based) alongside your ERPNext deployment.
2. **Configure** the MCP server with your ERPNext credentials and define the actions you want the LLM to access (e.g., create_sales_invoice, search_item).
3. **Connect** an LLM (OpenAI, Anthropic, etc.) by pointing it to the MCP endpoint; the model can now issue structured calls that the core translates into ERPNext API requests.
4. **Iterate** by adding custom actions or extending the protocol schema, then expose the service via Docker/Kubernetes for scaling.

**Production Readiness**

• **Activity & community**: 254 ★, 164 forks, recent commits (as of 2026‑06‑23), and a growing set of topics indicate an active codebase.
• **Maturity**: The core implements a clear API surface, includes SDK/CLI tools, and follows the MCP spec, making it suitable for pilot projects and gradual rollout to production.
• **Risks to verify**: Final checks on licensing (MIT/Apache?), security hardening of the MCP endpoint, and confirmation of long‑term maintainers are recommended before full‑scale deployment.

Overall, Frappe Assistant Core offers a high‑readiness, standards‑based way to embed AI assistants into ERPNext environments, enabling fast experimentation and scalable production use cases.

Problem

buildswithpaul/Frappe_Assistant_Core helps connect AI assistants to real tools and data through a standard protocol.

Outcome

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

Workflow

• connect AI agents to tools
• ship Model Context Protocol servers
• standardize integrations

Evidence

Infrastructure that connects LLMs to ERPNext. Frappe Assistant Core works with the Model Context Protocol (MCP) to expose ERPNext functionality to any compatible Language Model

Source