# aidesignblueprint/integrations [Health: Active]

**Category:** 🧠 Knowledge & Memory  
**Repository:** https://github.com/aidesignblueprint/integrations  
**GitHub Stars:** 3  
**Views:** 3  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/aidesignblueprint-integrations

## Description
Read-only doctrine access for the Agentic AI Blueprint — the industry standard reference for safe, observable, and steerable AI agent UX. Browse and search 10 principles, clusters, curated implementation examples, and application guides for production agentic AI systems. 13 public tools require no credentials. https://aidesignblueprint.com/mcp

## Claude Desktop Quick Installation
Remote MCP endpoint (confidence: high). Install path detected from listing signals. Add as a URL/SSE server in your client:

```json
"mcpServers": {
  "integrations": {
    "url": "https://aidesignblueprint.com/mcp"
  }
}
```

## Documentation

## What aidesignblueprint/integrations MCP server does

aidesignblueprint/integrations MCP server provides agent access to the public portion of the AI Design Blueprint. The content covers 10 principles, related clusters, curated implementation examples, application guides, and reusable assets for production agentic AI systems. Retrieval is read-only and anonymous, so an agent can consult the doctrine without providing credentials.

The repository also contains integration material for MCP, IDE rules, prompt files, agent runtimes, open-weight workflows, and structured doctrine exports. The MCP endpoint is the hosted public contract at `https://aidesignblueprint.com/mcp`.

## How it works

Configure an MCP client to initialize against the hosted endpoint. A basic verification sequence is to call `clusters.list()`, followed by an example search such as `examples.search(query="orchestration visibility steering", limit=3)`. Search parameters vary by resource: principle and guide searches accept a query and optional limit, while example searches can also filter by principle IDs, difficulty, and library.

The public retrieval surface does not require an account. Two opt-in signal methods are also available. `signals.report` records a value moment only after a user clearly says something was useful, while `signals.feedback` is intended for explicit requests to leave feedback. These calls accept structured fields and are not meant to store prompts, code, file contents, or secrets.

## Setup and configuration

Use the setup guides in the repository's `docs/setup/` directory for client-specific instructions, or add the relevant MCP configuration to the client or repository where the agent runs. For the hosted integration, use the exact endpoint `https://aidesignblueprint.com/mcp`; no environment variables or credentials are required for the public retrieval tools.

The repository is organized into shared doctrine files, MCP configuration and notes, provider-specific instructions, static prompt packs, and structured exports. It does not provide a CLI installer, a public OpenAPI schema, or a separate public HTTP API contract beyond MCP and static assets.

## Tools and capabilities

The public retrieval tools exposed by the aidesignblueprint/integrations MCP server include:

- List and fetch principles and clusters.
- Search principles by query.
- Fetch and search curated implementation examples, with optional filters.
- List assets.
- List, fetch, and search application guides.
- Submit opt-in value signals or explicit qualitative feedback.

Additional authenticated tools include learning-path and coaching-context access, evidence submission, team summaries, and three validation surfaces. The validation tools cover agentic implementation quality, rendered frontend experience design, and written specification quality. They require protected access; the implementation, design, and specification validators are identified as Pro/Teams features.

## Limitations and notes

Anonymous access covers retrieval and the documented signal methods, not the protected tools. Signal calls must follow the stated consent rules: value reports should not be automatic or silent, and feedback should not be requested proactively. Do not put proprietary code, file contents, or secrets into signal context fields.

The repository is intended to mirror the canonical material on the AI Design Blueprint site. Changes should be checked against the site's MCP endpoint, agent documentation, and agent assets. The available material does not specify a repository license or a local package command for running the MCP server, so the hosted endpoint is the supported installation path described here.

_Full upstream README: https://allmcps.com/mcp/aidesignblueprint-integrations/readme_