Scope Guard

io.github.mdfifty50-boopPublished by GitHub account mdfifty50-boopv0.1.3Updated Oct 24, 2026

Hard scope boundaries to prevent AI agent drift

The publisher's own server: Scope (AEC) - Preview

VerifiedSTDIODesktop onlySecurity & MonitoringAI & ML

Overview

AI-generated overview

Local MCP server that defines scope contracts for AI agents and checks their tool calls, file access and delegation against those boundaries.

What you can do
Create and list scope contracts defining allowed and forbidden tools, file access patterns and delegation limits, and read active contracts as a resource. Validate a proposed tool call or file access against scope before it runs. Analyze delegation chains for circular patterns and detect agents duplicating the same work across action logs. Log agent actions and produce per-agent compliance reports with violations, delegation counts and scope utilization.
When to use it
Useful when several agents work together and you want explicit boundaries, pre-flight checks on tool and file use, and reports on drift, duplicated work or delegation loops.
Requirements
Runs as a local process over stdio, installed from the npm package scope-guard-mcp (for example via npx). No authentication, environment variables or headers are declared. Desktop client only; not executable on the web.
Before you install
It records agent actions and scope contracts locally, so logs may contain sensitive details of what agents did. Validation is advisory: it reports whether an action is in scope but does not itself block the call.

Installation

In SourceWeft

  1. Open Scope Guard in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.

Other MCP clients

Follow the launch instructions in the repository.

README

scope-guard-mcp

MCP server that enforces multi-agent scope boundaries. Prevents scope drift, work duplication, and infinite delegation loops in agentic systems.

Tools

ToolPurpose
create_scope_contractDefine agent boundaries: allowed/forbidden tools, file access patterns, delegation limits
validate_actionPre-flight check — is a tool call or file access within scope?
detect_delegation_loopAnalyze delegation chains for circular patterns
log_agent_actionTrack agent actions for compliance monitoring
get_compliance_reportPer-agent summary: actions, violations, delegation count, scope utilization %
detect_work_duplicationFind agents doing the same work across action logs

Resources

URIDescription
scope-guard://contractsAll active scope contracts

Install

bash
npm installnode src/index.js

Claude Desktop

json
{  "mcpServers": {    "scope-guard": {      "command": "npx",      "args": ["scope-guard-mcp"]    }  }}

License

MIT

Source: README.md at commit 5e78984

Tools

0
SourceWeft has not started this package in its sandbox yet, so its tools are not known.

Version history

1
  1. v0.1.3LatestSep 16, 2026

Score

40% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 days25 pts

    Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.

  • Auth declared10 pts

    Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.

  • Tools listed15 pts

    Answer tools/list without a credential, or ship a package that starts in a sandbox without one.

  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts

    Declare a license in the repository or the registry entry.

Other Scope integrations