Registry

com.policylayerPublished by policylayer.comOfficial publisherv1.0.1Updated Oct 24, 2026

The MCP server that vets MCP servers: identity, risk grade and per-tool risk before you install.

The publisher's own server: PolicyLayer

VerifiedStreamable HTTPWeb executableDeveloper ToolsSecurity & Monitoring

Overview

AI-generated overview

A hosted PolicyLayer service that lets an assistant look up published MCP servers and check their identity, risk grade and per-tool risk before installing them.

What you can do
Look up one MCP server by slug, npm package, URL or name and get its identity evidence, risk grade, auth posture and riskiest tools; unknown servers are queued for scanning by the lookup itself. Search the published registry by name, slug or package substring, with grade and verification on each match. Check a whole stack of up to 25 servers in one call and get a verdict. Inspect a single tool's risk category, severity, OWASP classes, parameters and recommended policy default. With a licence key, read the ordered change feed of drift, posture flips and impostor flags.
When to use it
Useful when an assistant is about to install or allow an MCP server and you want a published risk and identity record first, or when auditing an existing mcpServers configuration or deciding whether a specific tool should be permitted.
Requirements
A remote Streamable HTTP endpoint; nothing to install locally. Single-server lookups are free and need no key. The change feed tool requires a Registry Licence key sent as an Authorization Bearer header.
Before you install
The change feed needs a paid Registry Licence key, which is a credential to keep private. Lookups send the server identifiers you query to PolicyLayer's hosted service. The risk grades and recommendations are third-party assessments, not a guarantee of safety.

Installation

In SourceWeft

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

Web executable via Streamable HTTP. Remote servers run from the web runtime once configured in a workspace.

Use with an agent

Paste this prompt into Claude Code, Codex, Cursor or another agent — or into SourceWeft chat. The agent reads this server's install guide, shows you its publisher, endpoint and sign-in, and adds it to your client once you agree. Keys never go into a command.

Read https://sourceweft.com/mcp/com.policylayer%2Fregistry/install.md and connect the MCP server it describes to my client. Before changing any config, show me its publisher, endpoint and sign-in method and wait for my OK. Never put an API key in a command or in a file I would commit; ask me where it should come from.

Read the install guide the agent follows

Other MCP clients

Add this to your client's mcpServers config.

{
  "mcpServers": {
    "registry": {
      "type": "http",
      "url": "https://api.policylayer.com/mcp"
    }
  }
}

README

PolicyLayer Registry MCP Server

The MCP server that vets MCP servers.

Connect your agent to the PolicyLayer registry and it can ask one question before installing or allowing any MCP server: is this safe? The answer is the published registry record: verified identity with the evidence, an A to F risk grade, probed auth posture and every tool's risk classification.

  • Endpoint: https://api.policylayer.com/mcp (Streamable HTTP)
  • Free single-server lookups, no key needed
  • 32,500+ published servers and 515,000+ classified tools, kept current by continuous scanning
  • Docs: policylayer.com/registry/mcp

Connect

Claude Code

sh
claude mcp add --transport http policylayer https://api.policylayer.com/mcp

Claude (desktop / web)

Settings → Connectors → Add custom connector → URL https://api.policylayer.com/mcp

Cursor / Windsurf / VS Code

json
{ "mcpServers": {  "policylayer": {    "url": "https://api.policylayer.com/mcp"} } }

Then tell your agent: "check any MCP server against PolicyLayer before installing it." Add it to your team's agent instructions and every install gets vetted.

Tools

ToolWhat it returnsTier
check_mcp_serverFull record for one server: identity + evidence, risk grade, posture, tools riskiest-first. Unknown servers are queued for scanning by the lookup itself.Free
search_registryPublished servers matching a name, slug or package substring, with grade and verification on every match.Free
check_toolOne tool's full classification: risk analysis and evidence, OWASP classes, parameters, recommended policy default.Free
get_change_eventsThe ordered change feed: drift, posture flips, impostor flags. Cursor-based, with severity filters.Licence

Example

Ask your agent "is the github MCP server safe to install?" and it calls:

json
{ "name": "check_mcp_server", "arguments": { "server": "github" } }

and gets back the published record (trimmed):

json
{  "slug": "github",  "riskGrade": "D",  "identity": { "verification": "official", "confidence": "verified" },  "posture": { "auth": "gated", "rateLimited": false },  "toolCount": 86,  "tools": [    { "name": "delete_file", "category": "Destructive", "severity": "High", "recommendedPolicy": { "action": "deny" } }  ]}

Things to ask your agent

  • "Is the github MCP server safe to install?"
  • "We're adding the notion and linear MCP servers this week. Vet both and summarise the risks."
  • "Which of the MCP servers in this project can delete data or move money?"
  • "Should I allow the execute_sql tool on the postgres server? What policy does PolicyLayer recommend?"
  • "Find a verified MCP server for Slack and show me its risk grade."
  • "Check every server in my mcpServers config against PolicyLayer and flag anything below grade C."
  • With a licence key: "What changed in the servers we depend on since yesterday? Any posture flips or impostor flags?"

Adding a licence key

get_change_events needs a Registry Licence key (plr_...), self-serve at policylayer.com/registry/pricing.

  • Claude Code: append --header "Authorization: Bearer plr_..." to the command above
  • Cursor / Windsurf / VS Code: add "headers": { "Authorization": "Bearer plr_..." } inside the server entry
  • Claude desktop/web connectors can't send headers; free lookups only

Free and licensed

Everything about a single server is free and complete: the full record, every tool, the whole classification. A Registry Licence adds breadth across the catalogue: the change feed above, plus bulk snapshots, keyset paging and webhooks on the /v1 API.

What this repository is

The connect kit for a hosted service. The server runs at api.policylayer.com; there is no code to run from here. server.json is the listing published to the official MCP Registry as com.policylayer/registry. The MIT licence covers the contents of this repository only; the hosted service and the registry data are licensed separately.

PolicyLayer · Registry lookup · API reference · Licensing

Source: README.md at commit d95b8a6

Tools

5

check_mcp_server

Check an MCP server against the PolicyLayer registry BEFORE installing or allowing it. Accepts a registry slug, an npm package name (scoped or not), a remote server URL (https://…), or a server name. Returns the full published record: identity verification with its evidence, risk grade, auth posture, freshness, and the tool surface listed riskiest-first. A server the registry does not know is queued for scanning by this very call — check back shortly.

Parametersserverrequired

check_mcp_stack

Check a whole MCP stack against the PolicyLayer registry in one call — up to 25 servers, each given as candidate identifiers (npm package name, registry slug, or remote URL) tried in order until one resolves. Returns the published record for every hit — plus a deterministic verdict (attention signals and a suggested action) — and the lookup status for every miss; counts, grades and flagged tools come from the published records only. Costs one rate-limit unit per server.

Parametersserversrequired

search_registry

Search the PolicyLayer registry of published MCP servers by name, slug or package substring. Returns candidate matches with risk grade, identity confidence (verified / unverified / mismatch — mismatch means it claims to be an official server with no verifiable link to the brand) and tool count — follow up with check_mcp_server on the match you meant.

Parameterslimitqueryrequired

check_tool

One tool's full risk classification on a published MCP server: category, severity, risk analysis and evidence, OWASP classes, parameter schema and the recommended policy default. Use when deciding whether to allow a specific tool call, e.g. "should execute_sql on this server be permitted?"

Parameterstoolrequiredserverrequired

get_change_events

The registry change feed: tool-surface drift, auth-posture flips, impostor flags, version bumps — every event the freshness watchers emit, id-cursored so a consumer resumes exactly where it stopped. Requires a Registry Licence key (Authorization: Bearer plr_...); self-serve at https://policylayer.com/registry/pricing.

Parameterslimitafter_idseverity

Version history

1
  1. v1.0.1LatestSep 16, 2026

Score

100% 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 daysRequired25 pts
  • Auth declaredRequired10 pts
  • Tools listedRequired15 pts
  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts

Other Policylayer integrations

PolicyLayer

com.policylayerOfficial publisher

PolicyLayer is a remote MCP endpoint that acts as a system of record for AI agent authority, holding playbooks, routed policy questions and reusable rules.

Streamable HTTPWeb executableAuth
Security & MonitoringCredential requiredOct 25, 2026