ChainMemory
Portable, verifiable memory for AI agents — works across ChatGPT, Claude, Gemini and any MCP client
FreeAI AgentsMCPBrowser Extension1,521 impressions#16 of its week1 comment
Comments
>log in to comment- PDFops· 3mo ago
The Merkle-anchored integrity angle is the right kind of paranoia — but in my experience the harder failure in shared agent memory isn't tampering, it's two agents each writing a "current state" that's internally valid and mutually stale. Curious how consolidation handles that: last-writer-wins, a CRDT-ish merge, or per-agent branches reconciled on read? That design choice ends up mattering more than the proofs.
ChainMemory provides portable, verifiable, persistent memory for AI agents across models via Chrome extension, MCP server, or REST API.
- for
- Developers building AI agents that need cross-model context and audit-ready memory.
- pricing
- freemium
Key features
- Cross-model portability — Save memories in one model (e.g., ChatGPT) and recall them in others like Claude or Gemini.
- Cryptographic verification — Each memory gets a hash and project states are anchored with Merkle proofs for immutable integrity.
- Project Brain — Aggregates memories into a structured, queryable project state with decisions, milestones and risks.
- Multi-agent native support — Multiple agents share a single consolidated state with full attribution.
- Chrome extension & MCP server — One-click saving from chat windows or direct integration via MCP for any client.
- REST API & SDKs — Full API and Node.js/Python/Go SDKs to save, search, inject context and verify states programmatically.
Use cases
- Persist project decisions across ChatGPT, Claude and Gemini sessions
- Provide auditable AI decision trails for regulators or reviewers
- Share a common project state among multiple autonomous agents in a team
ChainMemory vs alternatives
| Best for | Cross-model verifiable AI memory | LLM app framework | Multi-agent workflow platform | LLM data indexing | MCP server discovery |
|---|---|---|---|---|---|
| Pricing | Freemium | Free | Subscription | Free | Free |
| DevHunt upvotes | 1 | 0 | 0 | 0 | 0 |
| Launched | Jul 2026 | — | — | — | — |
- ChainMemory vs LangChain: LangChain is a developer framework for building LLM apps, not a persistent memory service.
- ChainMemory vs CrewAI: CrewAI focuses on orchestrating multi-agent workflows, whereas ChainMemory stores and verifies shared memory.
- ChainMemory vs LlamaIndex: LlamaIndex provides data indexing for LLMs, not cross-model verifiable memory with blockchain anchoring.
- ChainMemory vs Smithery: Smithery is a registry for MCP servers; ChainMemory is an MCP-compatible memory layer with verification.
ChainMemory FAQ
Are my memories really owned by me?+
Yes, memories are linked to your API key, can be exported, migrated or deleted, and are never sold to third parties.
Does ChainMemory work with any AI model?+
It is model-agnostic and works with ChatGPT, Claude, Gemini, Perplexity, Copilot and any MCP-compatible client.
What is a “state fingerprint”?+
Each time a project state changes, a cryptographic hash is generated, enabling immutable audit trails and verification.
What happens to my data if I cancel my subscription?+
Your memories remain accessible under the Free tier limits; no data is deleted and accumulated credits stay in your account.
What are AIC credits?+
Plan credits are used for monthly operations (injects, anchoring, audits); “gift” credits accumulate month-to-month and never expire.
Summarized by DevHunt from chainmemory.ai · Oct 5, 2026. Details may change; check the official site.