blink artifact

Install design system

system_install

Destructive

Install a design system into an artifact you can edit (slides, design, dashboard or html): copies the system's tokens, README, fonts and images into the artifact; call again later to pull updates. system is the design-system id, or null to remove the installed copy. The copy is read-only: get_artifact lists it as _system/tokens.json and _system/README.md (html also _system/tokens.css) and reports system.status (current, outdated, unavailable). Returns { id, rev, system: { id, title, rev, status } | null }.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "id": "<id>",
      "system": "<system>"
    },
    "name": "system_install"
  }
}

Input Schema

Name	Required	Description	Default
`id`	Yes
`system`	Yes

Schema Changelog

Changes observed during successful MCP inspections.

First observedOct 29, 2026

TDQS

A4.1/5.0

Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations declare destructiveHint=true, and the description supports that by explaining that null removes the installed copy and that the installed copy is read-only. It also discloses where the files surface in the artifact (_system/tokens.json, README.md, tokens.css) and the status values, adding substantial context beyond the annotations. It does not state whether installing a different system overwrites the existing one or whether removal deletes files permanently.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Dense but front-loaded: the primary action and copy contents come first, followed by re-invocation and removal semantics, then artifact layout and return shape. It is a single long block rather than separated sentences, but no clause is wasted.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a destructive two-parameter mutation with no output schema, the description supplies the return shape ({ id, rev, system ... | null }), where the installed files appear, and the meaning of system.status. Remaining gap is the unstated nature of `id` and overwrite behavior on reinstall.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 0%, so the description must carry the load: it fully documents `system` as a design-system id or null-to-remove. However, the `id` parameter is never explicitly defined — the artifact id must be inferred from context — leaving half the required parameters unspecified.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb (install) and resource (design system) plus the target artifact types (slides, design, dashboard, html) and exactly what gets copied (tokens, README, fonts, images). This is clearly distinguishable from siblings like systems_list, system_default, and system_font_import.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Explicitly says 'call again later to pull updates' and that passing null removes the installed copy, so the re-invocation and uninstall semantics are clear. It does not name sibling alternatives (e.g., system_default or systems_list) for discovering or selecting a system, so it stops short of full when/when-not guidance.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All blink artifact tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources

Need Help?