// SCAN REPORT

AI-Canvas MCP Server

67 tools. 41 can modify or destroy data without limits.

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

Last updated: 30 Jun 2026

41 can modify or destroy data

26 read-only

67 tools total

Community server · catalogue entry verified 30/06/2026

How to control AI-Canvas MCP Server ↓

TOOL MAP

What AI-Canvas MCP Server exposes to your agents

Read (26) Write / Execute (38) Destructive / Financial (3)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS

Critical Risk

The most dangerous AI-Canvas MCP Server tools

Destructive · High delete_collection This tool performs an irreversible operation that destroys both the collection container and all contained items in a single action. Destructive · High delete_brand_profile This tool permanently removes a brand profile that likely serves multiple users or projects in a shared library. Destructive · High delete_collection_item This tool permanently removes data from a collection without the ability to undo the operation. Execute · High run_design_pipeline This tool runs a multi-stage pipeline that orchestrates other design operations (seen in sibling tools: apply_brand_profile, apply_operations, create_mockup).

41 of AI-Canvas MCP Server's 67 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control AI-Canvas MCP Server

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

Deny destructive operations

{
  "delete_brand_profile": {
    "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

{
  "apply_brand_profile": {
    "limits": [
      {
        "counter": "apply_brand_profile_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "audit_design_consistency": {
    "limits": [
      {
        "counter": "audit_design_consistency_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 AI-Canvas 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 AI-CANVAS →

Instant setup, no code required.

FULL CATALOGUE

All 67 AI-Canvas MCP Server tools

DESTRUCTIVE 3 tools

Destructive delete_brand_profile Delete a brand profile from the global library Destructive delete_collection Delete an entire collection and all its items from the project Destructive delete_collection_item Delete a single item from a collection

EXECUTE 3 tools

Execute generate_react Generate React components from design nodes Execute push_to_github Generate React components from design nodes and push them as a Pull Request to GitHub Execute run_design_pipeline Execute a full design workflow from brand to mockup

WRITE 35 tools

Write apply_brand_profile Apply a saved brand profile to the current project — injects brand tokens and activates constraint validation Write apply_collection_item Insert a collected asset into a page, with optional color mapping to the current brand Write apply_operations Apply a batch of design operations (insert, update, delete, etc.) to the document Write collect_asset Collect a design node from the current document into a named collection for reuse across projects Write create_checkpoint Create a named checkpoint in the document history Write create_mockup Place design into a mockup frame Write create_poster_template Create a poster layout based on brand identity Write create_print_document Create a new print-ready document artboard Write create_seed_component Create a new component definition from a seed Write export_pdf Export design nodes to PDF documents Write export_png Export design nodes to PNG images Write export_pptx Export brand collateral or slides to a PowerPoint presentation Write generate_brand_book Generate a complete brand guidelines book document Write generate_brand_tokens Generate design tokens from brand identity metadata Write generate_collateral Generate a specific brand collateral item Write generate_collateral_kit Generate a full kit of brand collateral items Write generate_color_palette Generate a color palette based on harmonies and brand guidelines Write generate_component_spec Generate a technical specification for a component Write generate_design_pattern Generate a node tree from a design pattern Write generate_design_spec Generate a design specification document Write generate_style_guide Generate a visual style guide document Write generate_tailwind_config Generate a Tailwind CSS configuration from document design tokens Write move_collection_item Move a collection item from one collection to another (target created if missing) Write open_document Open a design document from a file path or create a new one Write redo Redo the last undone operation Write responsive_preview Generate responsive layout previews for different viewports Write restore_checkpoint Restore the document state from a specific checkpoint Write save_brand_profile Save a brand profile (colors, typography, spacing constraints) to the global library for reuse in any project Write save_document Save the current document to a file Write set_brand Update brand identity metadata in the document Write set_tokens Set or update design tokens in the document Write undo Undo the last operation Write update_brand_profile Update a global brand profile: name, description, version, tags, or constraints Write update_collection Update collection metadata: name, description, tags, or brand profile association Write update_collection_item Update a collection item: name, description, rating, or tags

READ 26 tools

Read audit_design_consistency Audit design nodes for consistency across the document Read check_brand_compliance Validate all nodes in the current document against a brand profile Read check_color_accessibility Check color contrast and accessibility against WCAG standards Read check_print_quality Validate design for print production (DPI, bleed, CMYK) Read generate_redline Generate redline measurements for handoff Read get_brand Get brand identity metadata from the document Read get_design_patterns Get available design patterns and layout templates Read get_document Get the complete current document structure Read get_guidelines Get design guidelines and rules for a specific topic (web-app, landing-page, design-system, etc.) Read get_history Get the full operation history of the document Read get_layout Get the computed layout rectangles for nodes in the document Read get_print_presets Get standard print document presets Read get_screenshot Get a visual screenshot of a specific node or the whole page Read get_seed_components Get list of available seed components from the library Read get_style_guide Get aesthetic DNA and style guide based on tags or a specific style name (apple, linear, stripe) Read get_style_guide_tags List all available style guide tags for inspiration Read get_tokens Get design tokens defined in the document Read list_brand_profiles List all saved brand profiles in the global library Read list_checkpoints List all checkpoints in the document history Read list_collateral_types List available brand collateral types (business cards, stationery, etc.) Read list_collections List all asset collections in the current project Read list_mockups List available device and product mockup templates Read query_nodes Search for nodes in the document using semantic or text queries Read search_collection_items Search for collected assets across project and global library by name, type, tags, or brand profile Read validate_brand_compliance Validate if design nodes comply with brand identity guidelines Read validate_design_system Deep validation of design system components and tokens

EXPLORE

Related servers

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

FAQ

Questions about AI-Canvas MCP Server

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

Yes. The AI-Canvas MCP Server server exposes 3 destructive tools including delete_brand_profile, delete_collection, delete_collection_item. 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 AI-Canvas MCP Server? +

The AI-Canvas MCP Server server has 35 write tools including apply_brand_profile, apply_collection_item, apply_operations. 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 AI-Canvas MCP Server.

How many tools does the AI-Canvas MCP Server MCP server expose? +

67 tools across 4 categories: Destructive, Execute, Read, Write. 26 are read-only. 41 can modify, create, or delete data.

How do I enforce a policy on AI-Canvas MCP Server? +

Register the AI-Canvas 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 AI-Canvas MCP Server tool call.

Deterministic rules across all 67 AI-Canvas MCP Server tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON AI-CANVAS →

Instant setup, no code required.

67 AI-Canvas MCP Server tools catalogued and risk-classified — across an index of 43,000+ MCP servers.