MemoryGuard
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@MemoryGuardshow memory history for the last hour"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Your agents can write freely. MemoryGuard classifies, deduplicates, supersedes, quarantines, and compresses shared memory on every write - then lets you inspect, correct, or roll back the result afterward.
No account. No server. No telemetry. Your memory stays local.
Why MemoryGuard
Persistent memory solves only half the problem. When several coding agents write into the same context, memory can become duplicated, stale, contradictory, or unsafe to reuse.
MemoryGuard is the local control layer between your coding agents and their shared memory:
Instead of | MemoryGuard gives you |
A growing pile of unreviewed notes | Write-time classification, deduplication, superseding, conflict detection, quarantine, derivation, and compression |
Approving every agent write by hand | Automatic writes; human review only when you need it |
Treating an overwrite as permanent | History, evidence, supersede chains, and rollback |
Sending project context to another service | A local MCP stdio server with local SQLite storage |
Related MCP server: GroundMemory
See the governance loop
Agent writes memory
-> MemoryGuard MCP write
-> auto-organize
classify · deduplicate · supersede · detect conflict · quarantine · compress
-> active shared memory
-> governance console
inspect · correct · merge · lock · restore · roll backThe console is not an approval queue. Agents keep moving; you govern the outcome with evidence when it matters.
Install in 60 seconds
pip install agent-memguardChoose the coding agent you use. Each command adds MemoryGuard as an MCP server and writes its instruction file.
# Claude Code
memoryguard source add . && python -m memoryguard.provider_adapters install claude
# Codex
memoryguard source add . && python -m memoryguard.provider_adapters install codex
# Cursor
memoryguard source add . && python -m memoryguard.provider_adapters install cursorThen restart your agent and verify the environment:
memoryguard doctor
memoryguard mcp-statusNeed a desktop window for the governance console?
pip install "agent-memguard[gui]"For explicit configuration and provider-specific behavior, see the Claude Code, Codex, and Cursor guides.
What you can govern
Signal | What you can do |
Duplicate or stale memory | See the supersede chain and restore the prior version if needed |
Conflicting memories | Surface the conflict and resolve it deliberately |
Secrets, tokens, or credentials | Quarantine them instead of leaving them in active shared memory |
A wrong governance decision | Inspect its history and roll the shared memory back to a version snapshot |
Multiple coding agents | Bind agents to a governed shared-memory group |
What MemoryGuard is - and isn't
MemoryGuard is a local MCP memory backend and governance console for coding agents. It provides a shared source of truth and organizes writes as they arrive.
It is not a cloud service, an account system, or a human gate that blocks every memory write. It also does not pretend every agent's native memory can be disabled: provider support is reported as redirected, observed, or unsupported where appropriate.
Architecture
Layer | Responsibility |
Evidence layer | Agent-native memory, files, documents, and external MCP descriptors; raw inputs, not governance truth |
Memory layer | MemoryGuard's local MCP shared-memory backend; the governed shared source of truth |
Governance layer | GUI and CLI for observation, evidence, corrections, and reversible changes |
Core surfaces
Surface | Use it for |
MCP memory backend | Read, search, write, update, delete, and inspect shared-memory status |
Auto-organizer | Classify, deduplicate, supersede, detect conflicts, quarantine, derive, and compress on write |
Governance console | Review raw writes, conflicts, quarantine, supersede chains, and versions |
Provider adapters | Set up Claude Code, Codex, or Cursor from one command |
CLI | Audit local sources, manage authorized inputs, inspect reports, and manage memory builds/releases |
The complete MCP tool reference and CLI command reference are below for evaluation and integration work.
Command | Description |
| Read-only scan; generate a report |
| Open the latest report in a window |
| Explain a finding's evidence and risk |
| Generate a minimal fix plan without writing |
| Apply a plan: backup, patch, and rescan |
| Rescan and compare before/after |
| Restore from backup and re-verify |
| Manage authorized sources |
| Read-only scan and coverage ledger |
| Preview or create an offline import bundle |
| Memory build, verify, and release rollback workflows |
| Diagnose installation and environment |
| Inspect MCP shared-memory status |
Group | Tools |
Memory backend |
|
Audit and scan |
|
Agent binding |
|
External MCP |
|
Document extraction |
|
Semantic and provider |
|
Privacy and safety boundaries
MemoryGuard runs as a local MCP stdio server; it requires no account, remote server, or telemetry.
Shared memory is stored locally in SQLite under
.memoryguard/.Source scanning is read-only by default. Changes use explicit plans, backups, rescans, and undo paths.
A quarantined memory is deliberately kept out of active shared memory until you decide what to do with it.
Roadmap
Now: local MCP backend, auto-organization, governance console, provider adapters, and rollback.
Later: enhanced governance signals such as decay, derivation, and governance reports. No committed date.
Later: team and enterprise capabilities after proven demand. No committed date.
Contributing
Issues and pull requests are welcome. Read CONTRIBUTING.md; by submitting a pull request, you agree to the CLA.
License
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Latest Blog Posts
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/irisxc4/memoryguard'
If you have feedback or need assistance with the MCP directory API, please join our Discord server