Your agent builds the
blockchain data API you need
Your coding agent builds, tests and deploys an Envio indexer from one prompt. Then your apps and agents query the API it serves, in real time.
- to index 400,000 wstETH events on Monad, with the indexer deployed by an agent from one prompt
- ~20s
- Agentic indexing with Envio Cloud
Agents on both sides of the API
Your AI coding agent builds and runs the indexer. Your apps and AI agents query its data.
- Agents build it
From one prompt to a live API
envio init detects the agent, the built-in skills guide its config, schema and handlers, its tests check the work, and the envio-cloud CLI deploys it.
Learn more → - Agents use it
An API your agents can act on
Your agents query a GraphQL API kept current block by block, with reorgs handled, instead of raw RPC calls.
Learn more →
Everything an agent needs, in the terminal
Envio is set up for AI coding agents from the first command to the deploy.
An agent-aware CLI
envio init detects your agent and gives it the options and next steps.
17 agent skills built in
Every new indexer includes a skills folder. envio skills update keeps it current.
Agents check their own work
The agent writes a failing test, then the handler, and checks the result with tests.
Docs on tap
The docs MCP server lets agents search and read the docs. So does envio tools, with no setup.
Docs written for models
llms.txt, AGENTS.md and a markdown copy of every docs page.
Deploys from the terminal
The envio-cloud CLI deploys indexers, watches them sync and reads logs.
Your agent migrates the indexer for you
Point your coding agent at an existing subgraph or Ponder project and it ports it to HyperIndex, following the migration prompt in the docs.
Scaffold
Open your coding agent in an empty folder, in plan mode, and give it this prompt. Following the migration guide, it scaffolds one contract on one chain, then updates config.yaml to match your indexer's contracts, chains and events.
prompt Migrate my existing indexer at <EXISTING_INDEXER> (a local path or git URL) to Envio HyperIndex. Follow the guide at https://docs.envio.dev/docs/HyperIndex/migrate-with-ai phase by phase: scaffold, plan, implement, verify.Plan
It maps every part of the existing indexer to its HyperIndex equivalent and flags anything without a direct match, before it writes any handler code.
Port and verify
It ports one handler file at a time, runs codegen after each, then checks every item in the plan. For a subgraph, compare the two endpoints with the Indexer Migration Validator.
Give your agent the Envio docs
Add the docs MCP server to your agent, or search the docs from the CLI.
https://docs.envio.dev/mcp{
"mcpServers": {
"envio-docs": {
"url": "https://docs.envio.dev/mcp"
}
}
}envio tools search-docs "createTestIndexer"
envio skills updateGuides for building with agents
Walkthroughs of a coding agent building, deploying and migrating Envio indexers, and connecting it to the docs.
- Guide
Agentic blockchain indexing with Envio Cloud
An agent scaffolded, configured, pushed and deployed a wstETH indexer on Monad from one prompt.
Read the guide → - Guide
AI-assisted subgraph migration
Move a subgraph to HyperIndex with a coding agent, step by step.
Read the guide → - Guide
The Envio docs MCP server
Give any coding agent or MCP client the Envio docs, with search and full pages.
Read the guide →
Envio for coding agents FAQ
Which agents work, the tools they use, how they deploy and how they migrate an existing indexer.
Which coding agents work with Envio?
Any capable coding agent that can run commands and edit files. The CLI, the agent skills and the docs MCP server aren't tied to one agent.
Quickstart with AI →Can an agent migrate my existing indexer?
Yes. Migrate with AI gives your agent a phased prompt: it scaffolds the indexer, plans the migration of your subgraph or Ponder indexer, ports the handlers one file at a time, then verifies the result.
Migrate with AI →Let your agent build its first indexer
Open your coding agent, run pnpx envio init, and follow the Quickstart with AI.