Hybrid search, graph indexing, and vector-symbolic memory for AI agents
Open-source AI agent orchestration platform for incident response and automation
Stage
Startup
Not given
Founded
2026
Not given
Based in
US
Not given
Pricing model
Open source
Open source
Pricing
Open source under Apache-2.0 license. WARM Cloud is in development with usage-based pricing for sync, backup and verification; details to be published at launch.
Not given
Key features
Hybrid search combining BM25 lexical and vector scoring
Vertex graph vector index with 2.29-2.75x throughput vs hnswlib
Vector-symbolic fact memory with holographic reduced representations
AES-256-GCM encryption at rest with Argon2id key derivation
Signed audit log with Ed25519 and W3C DID identity
Secure purge that overwrites deleted data with zeros
Crash-safe append-only log with checksumming
Native Rust API with Python HTTP client and Node.js bindings
MCP server integration for editors and AI agents
WARM Cloud sync, backup and verification without server access to keys or documents
Incident response with agent investigation and pull request generation
Automated workflows triggered on schedules or webhooks
Agent summoning from GitHub, GitLab and Radicle with /ryph command
Semantic memory system for shared and persistent agent learning
Credentials kept behind gateways with no access in sandbox
Customized environments per repository with cached dependencies
Browser use for agent interactions with live watching
Artifacts and pages for reports and overviews
Deployment options from laptop to Kubernetes
Tool proxy with allowlisted MCP server integrations
What makes it different
Runs entirely locally in your process with no network calls or telemetry
Encryption at rest built into the open-source engine itself, not an enterprise add-on
Significantly faster vector search than hnswlib with graph-indexed vertex retrieval
Signed provenance and verified audit trail using C2PA, COSE and SCITT standards
Embedded library model eliminates dependency on external services or vendor lock-in
Agents work without holding real credentials or keys
Changes ship as patches that require review before merge
Runs on your own infrastructure from laptop to cloud
Semantic memory learns across runs with deduplication and contradiction handling
Built-in safety assuming agents can be manipulated or exploited
warm-rust
Ryph