Multi-agent AI fleet infrastructure with semantic search, temporal cueing, and conservation-driven orchestration. Every instance is a vessel. The fleet emerges from signal overlap.
# Install the client SDK
npm install @superinstance/tminus-clientimport { TMinusClient } from '@superinstance/tminus-client';
const client = new TMinusClient('https://fleet-vector-api.casey-digennaro.workers.dev');
// Semantic search across 1,500+ AI crates
const results = await client.search('distributed consensus protocol', 5);
console.log(results);Or just use cURL:
curl -X POST https://fleet-vector-api.casey-digennaro.workers.dev/search \
-H "Content-Type: application/json" \
-d '{"query": "multi-agent orchestration", "topK": 5}'📖 Full documentation → | API reference → | OpenAPI spec →
Every AI agent obeys a conservation law: γ + η = C
- γ (generation cost) — tokens, time, API calls
- η (innovation value) — new code, tests, discoveries
- C — total budget, conserved across all agents
Spend too much on generation (churning tokens without shipping), and η collapses. Ship too fast without thinking, and γ spikes on rework. The sweet spot isn't a setting — it's a law.
SuperInstance makes this observable.
| What | Where |
|---|---|
| Homepage | superinstance.ai |
| Semantic Search API | fleet-vector-api |
| npm Client | @superinstance/tminus-client |
| npm Dispatcher | @superinstance/tminus-dispatcher |
| Fleet MCP Server | fleet-mcp-server |
| Bottle Protocol | fleet-bottle |
| Conservation Tracker | fleet-conservation |
| Docs Hub | docs/index.html |
| Method | Endpoint | Description |
|---|---|---|
POST |
/search |
Semantic search across the crate ecosystem |
POST |
/recommend |
Context-aware crate recommendations |
POST |
/similar |
Find crates similar to a given one |
GET |
/stats |
Index stats (model, dimensions, crate count) |
GET |
/clusters |
Crates grouped by domain |
GET |
/docs |
Interactive HTML API documentation |
GET |
/openapi.json |
OpenAPI 3.1 specification |
Base URL: https://fleet-vector-api.casey-digennaro.workers.dev
Four vessels, heterogeneous hardware, one conservation law:
| Vessel | Hardware | Role |
|---|---|---|
| Oracle1 🔮 | Oracle Cloud ARM64 24GB | The Lighthouse — coordinates fleet, writes research |
| JetsonClaw1 ⚡ | Jetson Orin Nano | The Scout — edge inference, GPU-native rooms |
| Forgemaster ⚒️ | RTX 4050 WSL2 | The Forge — security audits, proofs, LoRA training |
| CoCapn-claw 🎭 | Kimi K2.5 / Telegram | The First Mate — frontend, collaboration, play-testing |
See CONTRIBUTING.md and GOOD_FIRST_ISSUES.md.
MIT