I genuinely find DDG more effective than Google Search. I have used it as a daily driver for 2+ years now. When i cannot find anything with DDG, i try with the !g macro, but results are more often than not even worse.
Super interesting idea! Congrats on the launch. Context is definitely something that is lacking in my experience. Im always frustrated when an agent cannot answer business-related questions, and i compare them to coding agents which seem to be able to answer everything. The difference is that coding agent has the context right there at the fingertips, while for business its gated behind a bunch of services and custom data models. Context is king :)
How do you handle encryption and confidentiality? Im building in this space too (MCP gateway https://www.gatana.ai/) which already have semantic search for tool outputs, and ensuring encryption and confidentiality is not trivial.
Extremely flexible and configurable MCP Gateway, target users is enterprises/companies/organizations who want secure and managed MCP within their company. Support both Cloud and On-premise.
Looks interesting. But how to explore or test or use? The product page (https://mistral.ai/products/forge) also does not contain anything useful. Just "Contact us"
Building Gatana, a platform for securely connecting an organizations agents to their services, with very flexible credential management and federated IDP trust.
Currently my mini-projects includes:
* 0% USA dependency, aim is 100% EU. Currently still using AWS SES for email-sending and GCP KMS for customer data key encryption for envelope encryption.
* Tool output compression, inspired by https://news.ycombinator.com/item?id=47193064 Added semantic search on top of this using a local model running on Hetzner. Next phase is making the entire chain envelop encrypted.
* "Firewall" for tool calls
* AI Sandboxes ("OpenClaw but secure") with the credential integration mentiond above
I implemented this as well successfully. Re structured data i transformed it from JSON into more "natural language". Also ended up using MiniLM-L6-v2. Will post GitHub link when i have packaged it independently (currently in main app code, want to extract into independent micro-service)
You wrote:
>A search for “review configuration” matches every JSON file with a review key.
Its good point, not sure how to de-rank the keys or to encode the "commonness" of those words
Very interesting, one big wrinkle with OP:s approach is exactly that, the structured responses are un-touched, which many tools return. Solution in OP as i understand it is the "execute" method. However, im building an MCP gateway, and such sandboxed execution isnt available (...yet), so your approach to this sounds very clever. Ill spend this day trying that out