XGuard — Agent Execution Gateway
io.github.moelayyan90/xguard-control-plane · v5.2.0 · score 89/100 · checked 2026-11-22
Automated, read-only signal from public data, not an audit or an accusation. Maintainers can dispute any finding. How grading works
Scoped agent execution. Server-side credentials, policy, budgets and verifiable receipts.
Source: https://github.com/moelayyan90/XGuard
Findings
| Severity | Rule | Detail |
|---|---|---|
| Low | MW005 | Remote endpoint lists 53 tool(s) to anyone, no credentials needed. |
| Medium | MW102 | 2 tool(s) can execute commands or delete data (execute_inference, execute_gpu_job). Require confirmation before the agent calls them. |
What was checked
- https://api.xguardgate.com/mcp: answers without auth
Public tools: xguard_search_offers, xguard_compare_suppliers, xguard_find_enterprise_software, xguard_find_gpu_capacity, xguard_request_quote, xguard_get_verified_facts, xguard_negotiate_offer, xguard_create_deal, xguard_purchase_service, xguard_verify_payment, xguard_get_receipt, xguard.action.catalog, xguard.action.quote, xguard.action.result, list_compute_providers, list_models, quote_inference, execute_inference, quote_gpu_job, execute_gpu_job, compute_health, get_execution_receipt, xguard.fact.get, xguard.authority.catalog, xguard.authority.command, xguard.authority.verify_receipt, xguard.commerce_burn_breaker, xguard.commerce_context_capsule, xguard.commerce_dom_to_api, xguard.commerce_prompt_firewall
Maintainer? Add the badge
Grades refresh daily. Think a finding is wrong? Dispute this grade.
[](https://www.mcpwatch.io/s/io-github-moelayyan90-xguard-control-plane-047be9e2)