The research behind the product
We open source the tools we build to solve hard problems in AI. These projects power what we do at Parslee—and they're free for anyone to use.
View on GitHubStateBench
How we verify AI memory actually works
AI systems often claim to remember things—but do they? StateBench is the testing framework we built to find out. It runs thousands of scenarios that expose common memory failures: facts that come back from the dead, information that leaks between conversations, corrections that get ignored. If an AI system passes StateBench, you can trust it to remember what matters.
- Tests for 6 distinct types of memory failure
- Over 1,400 real-world test scenarios
- Published methodology and research
Neo
The memory layer for AI coding tools
Every developer knows the frustration: your AI assistant forgets what you just told it, repeats the same mistakes, ignores your project's patterns. Neo fixes this. It's a reasoning engine that actually learns—building a knowledge base from every coding session that makes future suggestions smarter. Think of it as giving your AI assistant a memory that persists.
- Learns from both successes and failures
- Works with OpenAI, Anthropic, Google, or local models
- Runs locally—your code never leaves your machine
MCP-API
Turn any REST API into an AI tool
AI agents need to talk to APIs, but every API speaks a different language. MCP-API is the universal translator. Point it at any OpenAPI spec, Swagger doc, GraphQL schema, or Postman collection, and it instantly becomes a tool your AI can use. No custom integrations, no manual configuration—just connect and go.
- Supports OpenAPI 3.x, Swagger 2.0, GraphQL, and Postman
- Built-in auth handling (API keys, OAuth2, bearer tokens)
- Enterprise-grade security with Azure Key Vault encryption
Why we share our work
Building reliable AI is hard. By open sourcing our research, we help the whole industry move forward—and show that we understand these problems well enough to solve them for our customers.