Drop it in.
We'll remember.
Your operation's memory. The machines handle it. On your network. You oversee.
Every conversation, every briefing detail, every operational decision — captured, connected, and instantly retrievable. On your network. Under your control. Zero cloud dependencies. Your AI agents share the same brain — and coordinate themselves.
Memory infrastructure that never leaves your network.
Find anything in half a second
"What did the convoy commander report about route security last week?" — instant results with source, timestamp, and confidence score. Across all channels, all sources, searched at once. Under 500ms. Works air-gapped.
AI that connects the dots
Auto-extracts people, organizations, locations, dates, amounts, and operational terms. Builds a knowledge graph linking every interaction across every source. Fuzzy matching handles aliases and code names. Flags contradictions when reports conflict.
Your AI agents share one brain
Run multiple AI agents across analysis, monitoring, logistics, and comms? Through Agent Telepathy, they coordinate themselves — shared context, live broadcasts, intent signals. Zero conflicts, zero human coordination needed. All on your network.
Multi-source intelligence ingestion
Telegram, Signal, Gmail, Exchange, radio transcripts, field reports, document uploads — all captured on configurable schedules. Works with zero internet connectivity. Fully operational offline.
Air-gapped. No exfiltration possible.
Core operations require zero internet connectivity. No telemetry, no phone-home, no cloud APIs. Your data never leaves the box. Encrypted at rest with AES-256. API key auth on every endpoint — no OAuth middlemen.
Full audit trail
Every mutation logged with actor, timestamp, and reason. Immutable audit log exportable to your SIEM. Prometheus metrics endpoint. Health checks on all services. Compliance-ready by design.
Your AI agents just got a shared brain.
Your operation runs multiple AI agents — analysis, monitoring, logistics, comms. Now they coordinate through Kanister. Presence, broadcasts, shared context, coordinated work. On your network. Under your control. Your agents operate as a team. You oversee.
$ mb status
● agent-alpha active SIGINT processing
● agent-bravo active Logistics routing
● agent-charlie active Entity resolution
⟩ 3 agents, 0 conflicts, 47 entities today
Recent broadcasts:
→ sector-7: 3 new entities identified
→ convoy-alpha: route updated (secure)
→ context handed off: alpha → charlie
Network: air-gapped (no egress)
Signals: none active
Status: ALL CLEARSession Presence
Every agent registers a heartbeat. Who's alive, what they're focused on, when they last checked in. A standup that happens automatically.
Live Broadcasts
Agent detects new intelligence? Broadcasts it instantly. Everyone's in the loop on the next turn. Pub/sub for your AI workforce.
Shared Working Memory
Each agent writes what it's doing, what it decided, and what's pending. The next agent picks up exactly where the last one left off.
Intent Signals
Before touching critical resources, agents signal what they're about to change. Others work around it. Crashes auto-expire. No deadlocks.
Agent CLI
mb who — who's working. mb claim sector-7 — lock it. mb status — the full picture. You oversee. The machines work.
The Platform
12 MCP tools. Decisions API with semantic search. PM sync. Context assembly from Postgres + knowledge graph. Not a feature — the infrastructure your agents run on. All on your network, zero new dependencies.
Your agents are ready to work as a team. Give them the brain.
Request BriefingHere's what's under the hood.
Self-hosted. Air-gap ready. Open stack. No vendor lock-in.
55+ REST Endpoints
Full CRUD for memories, entities, sources, and ingestion. Hybrid search (vector + BM25 + graph). Agent coordination via /api/coordination/*. JSON responses, consistent error format, API key auth on every endpoint.
GET /api/search?q="route security sector 7"
POST /api/capture { content, source, metadata }
GET /api/stats
POST /api/coordination/sessions
POST /api/coordination/broadcasts
GET /api/coordination/signals?target=sector-712 MCP Tools
Structured tools for search, capture, entity management, coordination, and context delivery. Any MCP-speaking agent gets full memory access — Claude, custom agents, automated workflows.
mcp://kanister_search
mcp://kanister_capture
mcp://kanister_entity_lookup
mcp://telepathy_register
mcp://telepathy_broadcast
mcp://telepathy_checkTask Management + Decisions
Task API with claim/release and distributed locks. Decisions API — log every operational choice with semantic search. Context assembly from Postgres + knowledge graph in a single query.
POST /api/tasks { title, priority }
POST /api/tasks/claim { task_id, agent_id }
POST /api/decisions { topic, decision, rationale }
GET /api/decisions/search?q="route change"Air-Gap Ready Architecture
FastAPI backend. PostgreSQL 16 + pgvector. Redis 7 event bus. Memgraph knowledge graph. Qdrant vector store. Docker Compose — 10 containers, ~14.5 GB RAM. Core operations require zero internet connectivity. No telemetry, no phone-home.
The machines handle memory.
You oversee.
Self-hosted. Air-gapped. Agent-native. We work directly with defense and government teams to configure Kanister for your operational environment.
- Air-gap deployment — zero external network access
- Agent Telepathy — agents coordinate themselves on your network
- Agent OS — tasks, decisions, full audit trail
- 12 MCP tools + 55+ REST endpoints
- Full audit trail, SIEM export, immutable logging