# SpellVault’s evolution: Beyond LLM apps, towards the agentic future

[Grab](https://yomu.fyi/company/grab) · Felix Haryanto Lie · Nov 21, 2025

## Summary

Grab developed SpellVault as an internal no-code platform to democratize the creation of AI applications backed by Retrieval-Augmented Generation (RAG) and plugin integrations. To advance beyond static retrieval and linear input-output processing, the platform transitioned from its legacy executor to a graph-based execution model supporting branching, looping, and ReAct agent patterns. Capabilities like Python code execution and internal repository searching were unbundled from the prompt builder and consolidated alongside user plugins into unified Native and Community Built Tools. The platform also introduced a drag-and-drop deterministic workflow designer, automated task scheduling, and support for the Model Context Protocol (MCP).

## Takeaways

- Replacing the legacy linear executor with a graph-based executor transformed SpellVault applications into ReAct agents capable of branching, looping, and dynamic context gathering.
- Internal capabilities were structured into a unified Tools abstraction, distinguishing zero-configuration Native Tools from user-configured Community Built Tools.
- A drag-and-drop Workflow interface enables users to build deterministic sequences combining LLM queries, Python runtimes, and knowledge repository lookups.

**Tags:** [Architecture](https://yomu.fyi/topic/architecture), [Developer Experience](https://yomu.fyi/topic/developer-experience), [LLMs](https://yomu.fyi/topic/llm)

[Read original post](https://engineering.grab.com/spellvault-evolution-beyond-llm)