Single transactional memory layer
Stores documents, conversations, entities, attributes, relations, embeddings, and traces in one ACID-transactional database so memory stays inside a single substrate rather than split across multiple stores.
Spectron
Spectron is a provenance-first memory and knowledge layer for AI agents, built on SurrealDB. It is in early preview and uses a waitlist for access, with managed deployment available through SurrealDB Cloud.
What Spectron is
Spectron is a provenance-first, tri-temporal memory and knowledge layer for AI agents built on SurrealDB. It is designed to keep documents, conversations, entities, relations, embeddings, and traces in one ACID transaction so agent memory does not have to be stitched together from separate vector, graph, and row stores.
The product is presented as an early-preview system with a waitlist-based access model. The homepage frames Spectron as the memory layer, while SurrealDB Cloud provides the managed substrate underneath, with options for development, production, and enterprise deployment across the broader SurrealDB platform.
Core capabilities
Provenance and supersession
Captures provenance on every row and uses supersession instead of overwrite, so older beliefs remain visible and replacement is explicit rather than silent.
Multi-modal ingestion
Supports ingest of text, code, JSON, CSV, PDFs, images, audio, and video, with originals kept in object storage and structured records indexed in the database.
Structured extraction
Extracts typed entities, attributes, and relations from turns and document chunks, alongside embeddings, RAKE keywords, and location attributes.
Traceable retrieval
Uses a four-tier read path of direct lookup, response reuse, hybrid retrieval, and full-context fallback, with each read producing a retrieval trace.
Ongoing memory maintenance
Includes background reflection, elaboration, and consolidation to deepen the graph between interactions and to promote, update, or supersede facts with recorded provenance.
Where Spectron fits
Auditable agent memory
Build agent memory that can distinguish between conflicting facts, preserve superseded beliefs, and show which source supported a given answer.
Unified knowledge retrieval
Index documents, conversations, and multimodal artifacts together so an application can retrieve context without stitching across separate systems.
Shared team context
Support multi-user or multi-instance workflows where memory needs to be shared, scoped, and reconciled without relying on application-layer consistency tricks.
Managed platform evaluation
Use the managed SurrealDB Cloud path to evaluate the memory layer without provisioning and operating storage or infrastructure yourself.
Retrieval analysis
Debug agent responses by following the retrieval trace for a read and examining which tier, candidates, and rows were returned.
Pros and Cons
Pros
- Keeps memory, provenance, and retrieval state in one transactional substrate instead of across disconnected stores.
- Makes supersession and uncertainty explicit, which helps avoid silent overwrites and untraceable answers.
- Records retrieval traces for reads, which can support debugging and auditability.
- Supports both structured and unstructured inputs, including documents, conversations, files, and multimodal artefacts.
- Runs on SurrealDB Cloud, giving users a managed path to evaluate the platform without infrastructure setup.
Cons
- The page describes Spectron as an invite-only early preview, so it is not yet broadly available.
- The source material does not provide a standalone Spectron pricing page or a self-hosted Spectron deployment path.
- Feature detail is strong on architecture, but lighter on documented integrations beyond SurrealDB Cloud and the broader SurrealDB platform.
FAQ
Is Spectron generally available yet?
Spectron is presented as an early-preview memory and knowledge layer for AI agents. The page says the waitlist is the access path, with invites rolling out in weekly batches from launch week.
How do users get started with Spectron?
The page positions Spectron as the memory layer on top of SurrealDB Cloud, while the underlying SurrealDB substrate ships today. The homepage also points to a getting-started flow for Spectron on SurrealDB’s site.
What does Spectron do with incoming data?
Spectron’s workflow is described as ingest, extract, connect, and query. It ingests conversations and files, extracts typed entities and relations, reconciles beliefs with supersession and uncertainty, and exposes a retrieval trace on reads.
Does Spectron have separate pricing?
The source pages do not show a public standalone Spectron price. The pricing and cloud pages show SurrealDB Cloud free, scale, and enterprise offerings, and Spectron is described as running on SurrealDB Cloud.
Can Spectron be self-hosted?
The source highlights managed cloud deployment and a stateless application tier on top of SurrealDB. It does not document a self-hosted Spectron deployment path on the pages provided.
Quick Facts
- Category
- AI infrastructure
- Product type
- Agent memory and knowledge layer
- Platform
- SurrealDB / SurrealDB Cloud
- Access model
- Invite-only early preview with waitlist
- Source domain
- surrealdb.com
- Related pricing
- SurrealDB Cloud has free, scale, and enterprise plans
Spectron Alternativen
garden-md
garden-md is an open-source Node.js CLI that turns meeting transcripts into a local company wiki. It links entities across transcripts, keeps the original text intact, and renders the result as a browsable wiki.
Goldfish
Goldfish is an AI memory app for macOS that helps you reply, write, summarize, and continue using the context already on your Mac. It stores memory locally and writes in your tone.
Lasso
Lasso is an ecommerce product data platform for enriching catalog records, processing supplier files, generating product content, and monitoring competitors. It combines a web app with a REST API, SDK, and MCP server for teams and developers.
CreateOS Sandbox
CreateOS Sandbox is an isolated compute environment for running code and agent workloads inside Firecracker micro-VMs. It is designed for workflows that need machine-level isolation, private networking between sandboxes, and programmatic control through SDK, CLI, or MCP.
Codex Plugins
Codex Plugins bundle reusable skills, app integrations, and MCP servers into workflows you can install in the Codex app or use from Codex CLI. They help extend Codex with connected-service tasks, reusable instructions, and shared team workflows.
Struere
Struere is an AI-native platform for turning spreadsheet data into structured operational software with dashboards, alerts, and automations. It is aimed at teams that want to replace manual spreadsheet workflows without building custom tools from scratch.