| What it does | An engine for long-running agents that rebuilds its prompt under a cap and runs on your budget
| An HTTP document search API for your apps
|
|---|
| Stage | MVP
| Not given
|
|---|
| Founded | Not given
| Not given
|
|---|
| Based in | Not given
| US
|
|---|
| Pricing model | Freemium
| Open source
|
|---|
| Pricing | Prototype: free during beta, then from December 16, 2026 it costs $1 per 1,000 agent calls after 1,000 monthly included calls. Production: $20 a month with 20,000 agent calls included, then $1 per 1,000 calls after that. Top-ups: $1 per 1,000 calls, from $10 to $500 at a time, never expiring.
| Not given
|
|---|
| Key features | - Prompt rebuilt from journal under token cap at each step to keep it bounded as history grows
- Spending limits checked before each model call at token prices you declare
- Resume after crash without repeating finished actions
- Tool calls returned as actions for your code to carry out
- Proof of what ran: check prompts, re-run engine, download hash-chained journal
- Integrations with Vercel AI SDK, OpenAI Agents SDK, LangGraph, Mastra, Pydantic AI
- MCP server for Claude Code, Claude Projects and Codex
- Works with any model via bring-your-own-key
- TypeScript and Python clients for HTTP API access
| - Search passages that match a query with scores and file references
- Manage collections and add or remove documents over HTTP or web interface
- Share vaults and partitions for search-only access
- API keys with read, write or admin access levels
- Web interface for search and administration
- User account management with three access levels
- Full-text and vector search indexes
- Runs locally on 127.0.0.1 with no outgoing connections
- SHA-256 hashed API keys and scrypt hashed passwords
- Limit lists restrict collections and vaults per user or API key
|
|---|
| What makes it different | - Bounded prompt under a fixed cap while history grows, so costs per call stop climbing on long sessions
- Every run recorded with cryptographic proof: rebuilable prompts, hash-chained journal, re-runnable engine
- Resumes from crash without repeating finished actions using stable idempotency keys
- Budget enforcement before each model call, not reactive overage billing
- No model key storage or logging; your provider bills you directly
| - Runs entirely locally on your computer with no outgoing connections
- Free and open source with MIT license
- Does not write answers or contain a language model
- Shared vaults and partitions are search-only and cannot be modified
- Integrates with any language model of your choice
|
|---|
| Free plan | Not given
| Not given
|
|---|
| Open source | Not given
| Not given
|
|---|
| Platforms | Not given
| Not given
|
|---|
| Public API | Not given
| Not given
|
|---|
| Community votes | 0 | 0 |
|---|
| DR (Domain Rating by Ahrefs) | Not checked yet ● no change since last check | 1 ● no change since last check |
|---|
| Trust Flow (Majestic) | 0 | 3 |
|---|
| Citation Flow (Majestic) | 9 | 27 |
|---|
| Referring domains (Majestic) | 4 | 17 |
|---|
Argentic
ModelBrain RAG