[Proposal] Cloud memory for Venom users — 97.5% fewer tokens (ViBo API)

[vnbochkarev-netizen]

Hi Venom team,

Your vynect/venom is a SaaS product where AI agents talk to users — exactly where
persistent memory matters most. Without it, every session starts blank and
every request re-sends full history (expensive).

ViBo Cloud API — hosted memory for YOUR users, one integration:

• POST /memory/add — save facts per user (label, content, dedup)
• POST /memory/search — return ONLY relevant facts (97.5% fewer tokens)
• POST /memory/thread — compress long conversations (-72%)
• POST /subkey — you create a sub-key per your user (isolated memory,
  encrypted L1/L2/L3, secrets never reach the LLM)
• GET /usage — token savings stats

Zero infrastructure: we host, you integrate in 10 minutes (OpenAPI spec).
One master key, up to 100 user sub-keys on the Business plan ($20/mo).

For a support platform: each customer's history is instantly available —
agent replies in context, your LLM bill drops ~97%.

2-day free trial (100 facts): https://wwwvibo.com
Docs: https://github.com/vnbochkarev-netizen/ViBo-memory
OpenAPI: https://github.com/vnbochkarev-netizen/ViBo-memory/blob/main/integrations/vibo_openapi_dify.json

Open to integration help, co-marketing, or referral (30% recurring).
Honest: no savings on code gen — we publish our limits publicly.

Best,
ViBo team