by irisxc4
MemoryGuard is a local-first MCP memory server for coding agents. It organizes what agents write, keeps evidence for changes and lets governance decisions be rolled back.
PluginContact for pricing
Step 1: Follow the Quick start in the project README to install MemoryGuard locally.
Step 2: Register the local stdio MCP server with a supported host such as Claude Code, Codex, Cursor or TRAE.
Step 3: Let your agents write memory as usual; MemoryGuard organizes each write.
Step 4: Review conflicts and exceptions through the CLI or the desktop governance console.
Step 5: After upgrading, reconnect or restart your hosts and repair their provider integration.
No. The README says there is no account, no remote server and no remote telemetry.
Content that looks sensitive is quarantined from active memory.
Yes. Evidence and supersede chains are preserved and governance decisions are described as reversible, with rollback support.
The README names Claude Code, Codex, Cursor and TRAE.
Requirements
1Claw MCP1Claw MCP gives AI agents just-in-time access to secrets stored in a 1Claw vault, so keys are used without being pasted into prompts. It also includes tools for transaction signing and a security scanner for LLM content.
Phala CloudPhala Cloud runs Docker workloads, agents and private LLM inference inside hardware-backed trusted execution environments. It produces attestation reports so you can verify what ran.
Cloudflare AI GatewayCloudflare AI Gateway sits between your application and AI model providers. It adds analytics, logging, caching, rate limiting, routing with fallbacks and safety controls, and takes one line of code to start.