Persistent memory for AI agents.
Governed. Auditable. Yours.
Recall Racoon gives your AI agents memory you can defend. Claims are governed through their lifecycle and recalled with provenance. Projects stay scoped. Detected credentials move into an encrypted vault. Rumi organizes the incoming stream. Mission Control puts the Stash in view.
Capture once.
Recall when it matters.
Recall Racoon gives your AI systems persistent memory. Information enters from multiple sources. Rumi Rummages through it. Your Stash remembers. Your AI recalls.
Deterministic Control Plane
Separate reasoning from memory policy. Apply explicit rules to admission, conflict handling, and recall. Keep governance in the control plane.
Hybrid Semantic Search
Find stored knowledge by meaning and keywords. Hybrid retrieval brings relevant memories back with the context needed to use them.
Claim Lifecycle Governance
Candidate. Confirmed. Stale. Superseded. Conflicted. Archived. Track claims through their lifecycle with provenance, citations, and human review.
MCP-Native Integration
Connect compatible AI agents through MCP. Use REST for application workflows. Save and retrieve memory through the interfaces your systems use.
Local-First by Default
Run memory processing on your infrastructure. Use configured local models for embeddings and generation. Your environment. Your deployment choices.
Secret Redaction & Credential Vault
Detect supported secrets at write time. Route detected values to an encrypted credential vault and keep references in memory. If the vault is unavailable, redact detected values and preserve the remaining content.
Rumi handles the hustle.
Your organization is constantly producing new information. Agents discover things. Documents change. Projects evolve. Decisions are made. APIs send updates. People correct earlier assumptions. Rumi works through that incoming stream and helps turn it into organized, reusable memory.
Understand
Rumi examines incoming information and identifies what it is about.
Summarize
Important context can be condensed into useful memory without losing the original source.
Categorize
Rumi helps organize knowledge into meaningful categories and metadata.
Connect
Related memories can be associated so future Recall is not limited to exact words.
Prepare
Information is prepared for semantic retrieval and future use by connected systems.
Organize
The Stash becomes more than stored content. It becomes an organized memory layer.
Ask Rumi.
Need something from the Stash? Ask Rumi. Query stored knowledge, work with memory, and bring relevant context back into the conversation.
Rumi works in the background so your people and AI systems do not have to become memory librarians.
Rumi does not replace your AI tools. Rumi helps them remember.
New session. Same project.
Keep saved context available across sessions and restarts. Scope memory to the project. Manage it through Mission Control.
Session Continuity
New session. Same project. Persistent memory and namespaced session state let connected agents pick up saved context after a restart. Session expiry stays configurable.
Project-Scoped Memory
Separate projects. Scoped recall. Organize memory by user and project, with machine context where configured. Retrieve from the intended scope.
Mission Control
Your Stash, in view. Browse and edit memories. Manage sessions. Import and export knowledge. Explore connections through the memory graph.
Operational Visibility
Health checks. Usage metrics. Daily status reports. See service health and memory operations. Scheduled backups and configurable retention support recovery planning.
Experience the Stash
with synthetic data
Explore the Stash with synthetic data. Browse memories, search saved context, follow Rummage processing, and inspect the knowledge graph. This is a product simulation, not a connection to a production Stash.
Stash Browser
15 memoriesDeployment requires approval from SRE team via #deploy-requests channel. Runbook: wiki/deploy-v2
Staging environment auto-deploys on merge to main. Production requires manual gate.
Database migrations must run before service deploy. Order: migrate → smoke test → deploy.
Rollback procedure: revert commit, trigger pipeline, verify health checks.
Connection pool maxes at 20 by default. Increase via DATABASE_POOL_SIZE. Monitor: dashboard/db-health
Symptom: 503 errors with "no connections available". Cause: unclosed transactions in v2.3.1.
Fix: upgrade to v2.3.2 (connection leak patched). Verify: pgbouncer active.
Previous fix (v2.2.8) addressed different leak in connection factory. Not applicable.
Right to erasure: submit via privacy@company.com. SLA: 30 days. Verification required.
Data retention: customer data 7 years post-contract. Analytics data anonymized after 2 years.
ADR-0047: CQRS chosen for order domain due to conflicting read/write models.
Trade-off: eventual consistency acceptable (max 2s lag). Not acceptable for inventory—stays CRUD.
Previous attempt (ADR-0021): unified model caused N+1 on reporting. Reverted.
JWT tokens must never be logged. Redaction pattern: /[A-Za-z0-9-_]{20,}\.[A-Za-z0-9-_]{20,}/
AWS access keys detected in memory are auto-redacted at ingest. Alert sent to #security.
Dual-Plane Memory Architecture
The Data Plane stores memory and its context. The Control Plane governs admission, conflict handling, and memory policy. Reasoning and memory rules have distinct roles.
Agent Layer
Connected agents and applications
Transport
MCP and REST interfaces
Control Plane
Admission rules, policy enforcement, conflict review
Data Plane
Persistent memory, metadata, and search indexes
Rumi Governance
Claim validation, provenance, and review
Retrieval Engine
Relevant memories returned with context
Lifecycle Manager
Claim states, supersession, and audit records
Ready to deploy your own Stash?
Tell us about your agents, your infrastructure, and the knowledge they need to retain. We'll walk through the right deployment for your workflow.
Memory governed at every step.
Admission rules. Claim lifecycles. Source provenance. Recall Racoon separates reasoning from memory policy, with explicit controls over what enters the Stash and how it returns to your agents.
Control Plane
Enforce memory policies separately from model reasoning. Keep admission, conflict handling, and retrieval rules explicit.
Claim Lifecycle
Track candidate, confirmed, stale, superseded, conflicted, and archived claims. Keep changing knowledge under review.
Provenance & Audit Trails
Track sources and memory operations. Inspect the context behind a claim and the record of how memory changes.
Conflict Review
Surface conflicting information. Apply resolution policies. Bring people into the loop when review is needed.
Need a custom deployment?
Your agents. Your infrastructure. Your requirements. Discuss memory governance, integrations, and deployment boundaries with us.
More than stored context.
The Stash retains knowledge. Governance keeps its use explicit. Here’s what that means in practice.
| Capability | Recall Racoon |
|---|---|
| Claim lifecycle | Track claims as information changes. |
| Provenance | Keep source context and citations attached. |
| Conflict handling | Surface conflicting claims and route them through resolution or review. |
| Project scope | Retrieve from the intended namespace. |
| Credential handling | Separate detected secrets from ordinary memory. |
| Human oversight | Browse, edit, and manage stored knowledge. |
| Portability | Export memories and sessions as JSON. |
Common questions
No. The demo uses synthetic, precomputed content. It illustrates browsing, search, processing, and graph views without connecting visitors to a production Stash or credential vault.
Claims move through an explicit lifecycle. The control plane applies memory policies. Provenance and audit records make changes traceable. Conflicts can be routed for review.
Saved memories persist. Namespaced session state remains available until its configured expiry. Connected agents retrieve that stored context when work resumes.
When the credential vault is available, detected secrets are stored encrypted and replaced with references in memory. If the vault is unavailable, detected values are redacted. The remaining memory is preserved.
Yes. Recall Racoon supports configured local models for embedding and generation. Connected agents, optional integrations, and deployment settings determine the complete data flow.
Explore the interactive demo. Then tell us about your agents and deployment requirements. We’ll discuss the capabilities and next steps that fit your workflow.
Go deeper into
the Stash.
Product
Governed memory, hybrid retrieval, and Mission Control.
Rumi
The intelligence layer that Rummages through incoming content.
Architecture
How the control plane, memory store, and retrieval fit together.
Security
Credential handling, scoped access, and audit records.
Deployment
Cloud, private VPC, on-premise, and air-gapped options.
Enterprise
Governance for organizational AI workflows.
Use Cases
Engineering, support, and research agent workflows.
AI Memory Comparison
RAG, vector stores, and governed persistent memory.
Glossary
Definitions for the vocabulary of AI memory.
Research
How we think about evaluating memory quality.
Blog
Notes on governed memory, agents, and context.
Resources
Technical overviews of memory concepts.
Ready to talk to our team?
Contact SalesSales Inquiries
For demonstrations, deployment requirements, licensing, and project discussions.
sales@recallracoon.comSecurity & Governance
For questions about memory policies, credential handling, deployment boundaries, and audit records.
security@recallracoon.comWhat to expect
- A discussion of your agents and workflows
- A review of deployment requirements
- A walkthrough of relevant capabilities
- Next steps for your project
Private Deployment
Your infrastructure. Your memory. Discuss the environment, integrations, and operational requirements for your deployment.
Discuss Private DeploymentTell us about your project
All fields marked with * are required.