Fully open source. Runs in the browser. Bring your own Anthropic API key... nothing stored server-side.
We also ship an MCP server, so Claude Code, or any MCP-compatible agent can query the graph and consult the Oracle directly.
β https://t.co/P7rTWl24su
β https://t.co/57gFzcI3ST
Shoutout to @TheiaResearch for open-sourcing their list of 700 mental models.
At w3.wave we build agentic AI systems into our investment process, and we were using these frameworks internally but treating them as a flat list. The interesting question is how they actually relate to each other.
So we mapped and visualized it β https://t.co/i1rDHZvH2R π§΅
We also built the Oracle for anyone sitting with a hard decision.
Describe your situation. Fire it through the network. It runs a two-pass analysis across all 700 models and returns which frameworks are relevant, which ones challenge your direction, and a pointed question for each one to sit with.
That 60% stat is the unlock.
Built https://t.co/gkUqimMg8B after scanning hundreds of pages: great content, broken data layer (schema, identity, checkout).
AI checks your data before citing your content. If it fails, your answer-first pages never surface.
Kipp's right... but data trust matters as much as content structure.
Content + data trust = edge
@TheCorgiCompany is coming out of stealth with $108M raised to build the future of financial infrastructure.
A lot of the GDP flows through highly regulated financial entities that haven't been meaningfully changed in decades.
To fix this, we're starting in insurance.