// SCAN REPORT

PlugLayer MCP Server

61 tools. 26 can modify or destroy data without limits.

7 destructive tools with no built-in limits. Policy required.

Last updated: 30 Jun 2026

26 can modify or destroy data

35 read-only

61 tools total

Community server · catalogue entry verified 30/06/2026

How to control PlugLayer MCP Server ↓

TOOL MAP

What PlugLayer MCP Server exposes to your agents

Read (35) Write / Execute (19) Destructive / Financial (7)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS

Critical Risk

The most dangerous PlugLayer MCP Server tools

Destructive · High delete_app Deleting an app is an irreversible action that removes application infrastructure and data. Destructive · High delete_database Deleting a database is a destructive action that irreversibly removes data and cannot be undone. Destructive · High delete_deployment This tool removes/deletes deployed applications, which is an irreversible action that cannot be undone without restoration from backups. Destructive · High delete_project This tool irreversibly deletes an entire project, which cannot be undone.

26 of PlugLayer MCP Server's 61 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control PlugLayer MCP Server

PolicyLayer is an MCP gateway — it sits between your AI agents and PlugLayer MCP Server, and nothing reaches the server without passing your rules. These are the rules we recommend:

Deny destructive operations

{
  "delete_app": {
    "deny_if": [
      {
        "conditions": [],
        "on_deny": "Blocked by default. Requires approval."
      }
    ]
  }
}

Destructive tools should never be available to autonomous agents without human approval.

Rate limit write operations

{
  "add_custom_domain": {
    "limits": [
      {
        "counter": "add_custom_domain_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "analyze_compose_deploy_plan": {
    "limits": [
      {
        "counter": "analyze_compose_deploy_plan_per_minute",
        "window": "minute",
        "max": 60,
        "scope": "grant"
      }
    ]
  }
}

Controls API costs and prevents retry loops from exhausting upstream rate limits.

Create a free account and register PlugLayer MCP Server — nothing to install.
Add these rules — paste them, or build them visually. Tune the limits to your setup.
Point your MCP client (Claude, Cursor, anything) at your gateway URL.

ENFORCE POLICY ON PLUGLAYER →

Instant setup, no code required.

FULL CATALOGUE

All 61 PlugLayer MCP Server tools

DESTRUCTIVE 7 tools

Destructive delete_app Alias for remove_app() using app wording. Destructive delete_database Alias for remove_database(). Destructive delete_deployment Alias for remove_app() for clients still using deployment wording. Destructive delete_project Alias for remove_project(). Destructive remove_app Remove one of the authenticated user's apps. This deletes the runtime workload and revokes its active routing Destructive remove_database Remove one of the caller's provisioned databases, including its runtime workload and active routing. Destructive remove_project Remove one of the authenticated user's projects. PlugLayer removes the project's apps first, then tears down t

EXECUTE 10 tools

Execute deploy_compose Analyze docker-compose.yml, split it into separate deploy units, provision known databases through Data Layer, Execute deploy_image deploy_image Execute deploy_marketplace_template Deploy a marketplace template into an existing project or a brand-new project created during the same flow. Se Execute exec_app_terminal Run a shell command inside the user's own deployed app container and return the result. Uses a fixed 360-secon Execute redeploy Redeploy an existing app without changing its current slug. Confirm the exact app name with the user first and Execute restart_app Restart an app by queueing a redeploy. Default to recreate to optimize compute usage; use rolling only when lo Execute restart_database Restart a provisioned database by queueing its redeploy/restart flow. Execute sync_database_env_to_app Update one deployed app's env vars using a provisioned database's concrete connection details, then restart th Execute upload_image_archive_and_deploy Upload a locally built Docker/OCI image archive to PlugLayer. If the target app already exists in the project, Execute upload_image_archive_and_redeploy_app Rebuild flow for existing apps: upload a newly built image archive, push it with a new tag, keep the current s

WRITE 9 tools

Write add_custom_domain Add a custom domain. Confirm whether the user wants this custom domain or the default pluglayer.io subdomain f Write attach_custom_domain Attach a verified custom domain to an app. Set make_primary=true if you want it to become the app's main URL. Write create_database Provision a database from a ready marketplace template. The MCP tool resolves required DB names and secret-lik Write create_project Create a PlugLayer project namespace. Project creation only requires authentication. Write detach_custom_domain Detach a custom domain from its app while keeping verification. Write generate_github_actions Generate a GitHub Actions workflow YAML that builds an OCI archive and uploads it to PlugLayer for the same ap Write update_app_domain Update the app's default pluglayer.io route slug. Use this when the user chooses the built-in subdomain now an Write update_database_access Update a provisioned database's public TCP IP allowlist. Pass an empty list to allow all IPs again. Write update_user_context Update the caller's stored PlugLayer user context memory. Use this carefully for durable preferences, project/

READ 35 tools

Read analyze_compose_deploy_plan Analyze a docker-compose.yml and show how PlugLayer will split it into marketplace databases, separate compose Read check_database_slug_availability Check whether a database/Data Layer slug is available in a project before provisioning or renaming a database. Read check_slug_availability Check whether a PlugLayer slug is available inside a project before deploying or renaming an app/database. Read detect_custom_domain_provider Detect the likely DNS/domain provider for a custom domain from public NS records so the agent can confirm it w Read estimate_compute Estimate the compute needed for a described workload and return a tailored PlugLayer offer link. This is the p Read get_app_connection_env_vars Get env vars and connection fields for an app. Use this after provisioning so you can update dependent apps wi Read get_app_logs Alias for get_logs() using app terminology. Read get_apps_by_project get_apps_by_project Read get_compose_local_build_commands Analyze a docker-compose.yml and return exact docker buildx, smoke-test, and OCI-export commands for any local Read get_compute_summary Show accessible account-level compute. If the user is still planning capacity, prefer estimate_compute() first Read get_current_user Show the authenticated PlugLayer user and roles from Authentik. Read get_database_connection_details Get a provisioned database's connection strings, env vars, and docs so you can wire other apps automatically. Read get_database_logs Get recent logs from a provisioned database. Read get_deployment_status Get current app/deployment status and public URL. Read get_domains_by_project Alias for list_project_domains() using project-first wording. Use this before asking the user which domain the Read get_logs Get recent logs from an app. Read get_marketplace_template Get one marketplace template, including env var requirements. Use this before template deployment so you can r Read get_my_available_compute Show the current user's available compute in an end-user friendly format. When capacity is unclear, call estim Read get_my_available_computes Alias for get_my_available_compute(). When the user has not sized their workload yet, estimate_compute() shoul Read get_my_projects Alias for list_projects() using end-user wording. Read get_project Get project details for one of the authenticated user's projects. Read get_task_status Check async operation status, progress, result URL, or error. Read get_user_context Load the caller's stored PlugLayer user context memory. Use this at the start of a session when prior project/ Read inspect_local_github_repo Check whether the local repo exists, has git initialized, and points at a GitHub origin. Read list_database_templates List database-ready marketplace templates. Use this first when the user asks to provision a database through P Read list_deployments List apps. Optionally filter by project_id. Read list_marketplace_templates List deployable marketplace templates. Use this before deploying a ready-made template through PlugLayer. Read list_nodes List compute nodes accessible to the authenticated user. Read list_pluglayer_compute_options List PlugLayer marketplace compute options the user can buy. Use this after estimate_compute() when you want r Read list_project_domains List custom domains for a project and show their verification state. Read list_projects List the authenticated user's projects. Read list_registries List registry destinations available to the current user. Read list_user_databases List the authenticated user's databases, optionally filtered to one project. Read rollback Roll back an app to a previous revision. Read verify_custom_domain Verify the TXT and route DNS records for a custom domain. Use this after the user says the DNS records are add

EXPLORE

Related servers

Other MCP servers with similar tools — same risk classification, starter policies for each.

FAQ

Questions about PlugLayer MCP Server

Can an AI agent delete data through the PlugLayer MCP Server MCP server? +

Yes. The PlugLayer MCP Server server exposes 7 destructive tools including delete_app, delete_database, delete_deployment. These permanently remove resources with no undo. PolicyLayer blocks destructive tools by default so they never reach the upstream server.

How do I prevent bulk modifications through PlugLayer MCP Server? +

The PlugLayer MCP Server server has 9 write tools including add_custom_domain, attach_custom_domain, create_database. 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 PlugLayer MCP Server.

How many tools does the PlugLayer MCP Server MCP server expose? +

61 tools across 4 categories: Destructive, Execute, Read, Write. 35 are read-only. 26 can modify, create, or delete data.

How do I enforce a policy on PlugLayer MCP Server? +

Register the PlugLayer MCP Server 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.

Enforce policy on every PlugLayer MCP Server tool call.

Deterministic rules across all 61 PlugLayer MCP Server tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON PLUGLAYER →

Instant setup, no code required.

61 PlugLayer MCP Server tools catalogued and risk-classified — across an index of 43,000+ MCP servers.