$ loading_
Give agents memory with recall and provable knowledge deletion.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "obliviate" yet — see the docs or source repo.
Save the following user preferences to memory: reply in Chinese, prefer concise formatting, and the project codename is Aurora. Then confirm the key points that were stored.
The tool stores the preferences in memory and returns a summary of the saved key details.
Retrieve the saved information related to the Aurora project from memory and list it as bullet points.
The tool returns previously stored memory about the project so work can continue with context.
Delete all knowledge in memory related to the Aurora project codename and provide proof or a certificate of erasure.
The tool removes the related memory and returns a signed proof of erasure or equivalent deletion result.
Developers building MCP-based agents can use it to store user preferences, project context, and ongoing task details, then recall them in later interactions. This reduces repeated input and helps the agent maintain continuity.
When an agent has retained knowledge that should no longer be kept, this tool can perform provable deletion. Its focus on cascade deletion, cryptographic shredding, and signed erasure certificates fits scenarios that require deletion traceability.
Researchers or product teams testing agent behavior can use it to manage memory creation, retrieval, and removal in one place. This makes it easier to observe how the agent relies on memory during tasks.
It is an MCP server for agent memory that can remember, recall, and provably forget knowledge. The description also mentions cascade deletion, cryptographic shredding, and signed erasure certificates.
Based on the given description, it supports cascade deletion, cryptographic shredding, and signed erasure certificates. This suggests it focuses not only on deletion, but also on proving that deletion occurred.
The provided material does not include installation steps, runtime requirements, or key information. See the source repository.
Manage durable project memory in AGENTS.md for AI agents.
Give AI agents persistent, self-managing memory with recall and forgetting.
Give AI agents verifiable memory with citations, abstention, and compliant forgetting.
Store and retrieve agent lessons to improve tasks and avoid repeated mistakes.
Give AI agents long-term memory with user-scoped storage, recall, and deletion.
Provide self-hosted, encrypted, durable memory management for AI agents.