i built nuthatch, a blockchain indexer you run yourself.
one rust binary. no postgres, no docker, no api key.
nuthatch init 0xYourContract --chain mainnet
nuthatch dev
sql over your contract's events in under two minutes. MIT/Apache.
https://t.co/GVlz5owxZJ
timed nuthatch against ponder, envio and a subgraph on the same weth blocks and the same keyless rpc
first query back in 1.0s on nuthatch, 2.9s ponder, 26s envio, 189s graph-node
envio's hypersync path needs an account so i left it out. method and scripts are in the repo if you want to rerun it
To everyone building infra, dev tools, etc - how do you get agents to use your stuff? How do you market for agents? It's one thing when people coded by hand and searched in google and stackoverflow but quite another now
@ruzkypazzy not yet tbh. had claude code and codex do plain indexing tasks last night and neither found it, both just wrote their own getLogs loops
rerunning it as the agent docs land, curious if it moves
been running nuthatch on the graphops base archive rpc and its been great tbh
with blocks every 200ms you need full history that actually keeps up, and theirs does. if youre indexing base start here
https://t.co/I5OBr4m6cf
Base now produces a block every 200ms, about 10x the blocks it had before. That's a lot more history to index, and keeping all of it is getting more expensive.
We run Base archive RPC and full-history subgraphs, so your data goes back to genesis. DMs open if you're feeling the pain.
GrokBot team just dropped 1-hour workshop on running your work with AI agents from 0% to 100%:
“GrokBot is not a chatbot”
• 10% → 0:57 - how GrokBot team manage usage limits
• 30% → 6:41 - how to build skills for GrokBot
• 55% → 10:41 - building Chief of staff agent
• 80% → 12:48 - GrokBot orchestrator for Cursor agents
• 100% → 27:01 - how to replace your software with bots
most people still ask it questions like a chatbot - the team runs it as a teammate that directs other agents
watch it today - then read the full GrokBot playbook below ↓
Why would I use Grok Bot when I have Claude Code on a server with remote session in the Claude app? Don't get me wrong Grok Bot is interesting but it's kinda like an overpriced overhyped Hermes
Base now produces a block every 200ms, about 10x the blocks it had before. That's a lot more history to index, and keeping all of it is getting more expensive.
We run Base archive RPC and full-history subgraphs, so your data goes back to genesis. DMs open if you're feeling the pain.
@0xeabz like the self hosted history idea. on evm i do the same job with one rust binary and no postgres.
`nuthatch init` a contract, then `nuthatch sql` or `nuthatch mcp`
https://t.co/Z1bc99cRu8
i got tired of standing up postgres just to query a contract's events.
nuthatch is one rust binary. init a contract, then `nuthatch sql` or `nuthatch mcp`
if you already have an agent that speaks mcp, this is the local hook for contract history.
`nuthatch mcp` serves a nest as sql, no key and no rate limit once its indexed
@triport_io yea getLogs is the part that hurts on a public rpc. i drop a short window into a nest with `nuthatch dev --backfill 300` and query it as sql after that