38 tools. 25 can modify or destroy data without limits.
1 destructive tool with no built-in limits. Policy required.
Last updated:
Free to start. No card required.
Financial operations (payments_refund) can move real money. An agent caught in a loop could drain accounts before anyone notices.
Write operations (assess_legacy_codebase, audit_accessibility, figma_push_variables) modify state. Without rate limits, an agent can make hundreds of changes in seconds — faster than any human can review or revert.
Execute tools (forge_assess, forge_gate) trigger processes with side effects. Builds, notifications, workflows — all fired without throttling.
{
"payments_refund": {
"deny_if": [
{
"conditions": [],
"on_deny": "Requires human approval."
}
]
}
} Financial tools should be explicitly enabled per use case, not open by default.
{
"assess_legacy_codebase": {
"limits": [
{
"counter": "assess_legacy_codebase_per_hour",
"window": "hour",
"max": 30,
"scope": "grant"
}
]
}
} Prevents bulk unintended modifications from agents caught in loops.
{
"analyze_component_library": {
"limits": [
{
"counter": "analyze_component_library_per_minute",
"window": "minute",
"max": 60,
"scope": "grant"
}
]
}
} Controls API costs and prevents retry loops from exhausting upstream rate limits.
Yes. The Ui server exposes 1 financial tools including payments_refund. Without a policy, an autonomous agent can call these with no spend caps, no rate limits, and no approval flow. PolicyLayer lets you block financial tools by default, require human approval, or set per-tool rate limits — enforced on every call.
The Ui server has 22 write tools including assess_legacy_codebase, audit_accessibility, figma_push_variables. Set a rate limit in your policy -- for example, 10 calls per hour prevents an agent from making more than 10 modifications per hour. PolicyLayer enforces this at the gateway, before calls reach Ui.
38 tools across 4 categories: Execute, Financial, Read, Write. 13 are read-only. 25 can modify, create, or delete data.
Register the Ui MCP server in PolicyLayer, apply the suggested rules above (adjust the limits to your use case), and point your AI client at the PolicyLayer proxy URL instead of the server directly. Your agents keep the same tools; PolicyLayer evaluates every call against policy before it executes. Nothing to install, live in minutes.
Starter policies for each. Same risk classification, live on your fleet in minutes.
Deterministic rules across all 38 Ui tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
4,600+ MCP servers and 31,000+ tools scanned and risk-classified.