MELRA

Community
TypeScript

Local MCP execution runtime for agent tools: policy gates, scoped approvals, outcome verification, durable receipts, and recovery after interrupted calls.

About MELRA MCP Server

Local MCP execution runtime for agent tools: policy gates, scoped approvals, outcome verification, durable receipts, and recovery after interrupted calls.

Features

Tools
Getting Started

MELRA is currently an alpha release; APIs may change before 1.0. Try it in an isolated workspace, keep command and domain allowlists narrow, and review each consequential approval. Requires Node.js 22+. Run `npx @melra/cli@alpha setup` to write a safe starter policy, generate MCP client config, and run readiness checks. MELRA runs locally over stdio. See the repository docs for client-specific setup and security limits.

Tools (11)
Functions this server exposes to AI clients

melra_plan

Validate, persist, and apply policy to a bounded effect request.

melra_execute

Execute an approved plan, verify the declared outcome, and record the result.

melra_task_status

Read the durable state of a task.

melra_capabilities

Discover operations, platform support, limits and policy posture.

melra_task_cancel

Cooperatively cancel pending or running work.

melra_receipt

Retrieve redacted evidence and the execution certificate.

melra_workflow_plan

Validate, preflight, encrypt and persist a bounded workflow.

melra_workflow_advance

Execute one ready scheduling wave.

melra_workflow_status

Read the durable workflow projection.

melra_workflow_cancel

Cooperatively cancel nonterminal workflow work.

melra_workflow_control

Pause, resume or suspend a run without losing its place.

Compatible With
Claude Desktop
Cursor
VS Code