fix: remove mermaid blocks from pages, fix npm install, reorder sidebar

- Removed all 14 mermaid code blocks from docs pages (kept in llms.txt only)
- Fixed npm install flowmcp-core → npm install github:FlowMCP/flowmcp-core
- Moved Docs section below Roadmap in sidebar

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Author: a6b8

astro.config.mjs

@@ -97,6 +97,15 @@ export default defineConfig({
                         { label: 'Clients and Compatibility', translations: { de: 'Clients und Kompatibilitaet' }, slug: 'basics/clients' },
                     ],
                 },
+                {
+                    label: 'Roadmap',
+                    items: [
+                        { label: 'Overview', translations: { de: 'Uebersicht' }, slug: 'roadmap/overview' },
+                        { label: 'Integration', slug: 'roadmap/integration' },
+                        { label: 'Community Hub', slug: 'roadmap/community' },
+                        { label: 'Team', slug: 'roadmap/team' },
+                    ],
+                },
                 {
                     label: 'Docs',
                     translations: { de: 'Dokumentation' },
@@ -167,15 +176,6 @@ export default defineConfig({
                         },
                     ],
                 },
-                {
-                    label: 'Roadmap',
-                    items: [
-                        { label: 'Overview', translations: { de: 'Uebersicht' }, slug: 'roadmap/overview' },
-                        { label: 'Integration', slug: 'roadmap/integration' },
-                        { label: 'Community Hub', slug: 'roadmap/community' },
-                        { label: 'Team', slug: 'roadmap/team' },
-                    ],
-                },
             ],
         }),
     ],

public/llm/llms.txt

@@ -910,21 +910,6 @@ This page covers the practical guide for building agents. For the full specifica
 
 An agent manifest (`agent.mjs`) declares everything the agent needs: which tools to use, which model to target, how the agent should behave, and how to verify it works.
 
-```mermaid
-flowchart LR
-    A[agent.mjs] --> B[Tool Selection]
-    A --> C[Model Binding]
-    A --> D[System Prompt]
-    A --> E[Prompts]
-    A --> F[Skills]
-    A --> G[Resources]
-    A --> H[Tests]
-
-    B --> I["coingecko-com/tool/simplePrice"]
-    B --> J["etherscan-io/tool/getContractAbi"]
-    B --> K["defillama-com/tool/getProtocolTvl"]
-```
-
 ## Agent Manifest
 
 Each agent is defined by an `agent.mjs` file with `export const agent`:
@@ -2098,7 +2083,7 @@ node --version
 The core library provides schema validation, API execution, and MCP server activation.
 
 ```bash
-npm install flowmcp-core
+npm install github:FlowMCP/flowmcp-core
 ```
 
 ```javascript
@@ -2224,7 +2209,7 @@ For a new project using FlowMCP, a minimal `package.json` looks like this:
     "version": "1.0.0",
     "type": "module",
     "dependencies": {
-        "flowmcp-core": "latest",
+        "flowmcp-core": "github:FlowMCP/flowmcp-core",
         "@modelcontextprotocol/sdk": "latest"
     }
 }
@@ -2264,7 +2249,7 @@ Understand the architecture and data flow. See How It Works.
    mkdir my-flowmcp-project
    cd my-flowmcp-project
    npm init -y
-   npm install flowmcp-core
+   npm install github:FlowMCP/flowmcp-core
    ```
 
    Add `"type": "module"` to your `package.json` for ES module support.
@@ -2416,16 +2401,6 @@ FlowMCP is a **schema-driven normalization layer** that transforms any data sour
 
 No custom server code. No boilerplate. One schema per provider.
 
-```mermaid
-flowchart LR
-    E[REST API] --> A
-    F[SQLite DB] --> A
-    G[Workflow] --> A
-    A[Schema .mjs] --> B[FlowMCP Runtime]
-    B --> C[MCP Server]
-    C --> D[AI Agent]
-```
-
 ## Four Primitives
 
 FlowMCP v3.0.0 supports four primitives in a single schema file:
@@ -5265,13 +5240,6 @@ export const main = {
 
 The `origin` field determines where the runtime looks for the `.db` file:
 
-```mermaid
-flowchart TD
-    A["origin: 'global'"] --> B["~/.flowmcp/data/companies.db"]
-    C["origin: 'project'"] --> D[".flowmcp/data/companies.db"]
-    E["origin: 'inline'"] --> F["./data/companies.db (relative to schema)"]
-```
-
 | Origin | Path Resolution | Best For |
 |--------|----------------|----------|
 | `global` | `~/.flowmcp/data/{database}` | Shared datasets used across projects |
@@ -5413,16 +5381,6 @@ Skills are instructional multi-step workflows embedded in a schema. Unlike Promp
 
 How a skill is resolved and executed at runtime:
 
-```mermaid
-flowchart LR
-    A[User Input] --> B[Skill Content]
-    B --> C[Replace Placeholders]
-    C --> D[LLM Execution]
-    D --> E["Step 1: Call {{tool:X}}"]
-    E --> F["Step 2: Call {{tool:Y}}"]
-    F --> G[Structured Output]
-```
-
 The skill's `content` is a template. Placeholders like `{{input:tokenSymbol}}` are replaced with user-provided values, and `{{tool:computeRSI}}` references tell the LLM which tool to call at each step.
 
 ## Defining Skills
@@ -5598,19 +5556,6 @@ This page focuses on practical tool creation. See Schema Format for the full spe
 
 A schema file has two parts: the declarative `main` export that describes what tools do, and the optional `handlers` export that transforms requests and responses.
 
-```mermaid
-flowchart TD
-    A[export const main] --> B[namespace + name + root]
-    A --> C[tools]
-    C --> D[method + path]
-    C --> E[parameters]
-    C --> F[description]
-
-    A --> G[export const handlers]
-    G --> H[preRequest]
-    G --> I[postRequest]
-```
-
 ## The `main` Export
 
 The `main` export is a static, JSON-serializable object. No functions, no dynamic values, no imports.
@@ -5903,15 +5848,6 @@ npm install -g flowmcp
 
 The CLI follows a three-step pattern: discover tools, activate them, then call them.
 
-```mermaid
-flowchart LR
-    A[flowmcp search] --> B[Find tools]
-    B --> C[flowmcp add]
-    C --> D[Activate tool]
-    D --> E[flowmcp call]
-    E --> F[Execute tool]
-```
-
 ## Core Commands
 
 | Command | Description |
@@ -6034,18 +5970,6 @@ Instead of writing custom server code for each API, you declare schemas and Flow
 
 How FlowMCP bridges AI clients and APIs:
 
-```mermaid
-flowchart LR
-    A[AI Client] -->|MCP Protocol| B[FlowMCP MCP Server]
-    B --> C[Schema 1: CoinGecko]
-    B --> D[Schema 2: Etherscan]
-    B --> E[Schema 3: DeFi Llama]
-
-    C --> F[CoinGecko API]
-    D --> G[Etherscan API]
-    E --> H[DeFi Llama API]
-```
-
 The AI client sends tool calls over the MCP protocol. FlowMCP resolves the correct schema, validates parameters, calls the upstream API, and returns the result.
 
 ## Starting the Server

public/llms.txt

@@ -910,21 +910,6 @@ This page covers the practical guide for building agents. For the full specifica
 
 An agent manifest (`agent.mjs`) declares everything the agent needs: which tools to use, which model to target, how the agent should behave, and how to verify it works.
 
-```mermaid
-flowchart LR
-    A[agent.mjs] --> B[Tool Selection]
-    A --> C[Model Binding]
-    A --> D[System Prompt]
-    A --> E[Prompts]
-    A --> F[Skills]
-    A --> G[Resources]
-    A --> H[Tests]
-
-    B --> I["coingecko-com/tool/simplePrice"]
-    B --> J["etherscan-io/tool/getContractAbi"]
-    B --> K["defillama-com/tool/getProtocolTvl"]
-```
-
 ## Agent Manifest
 
 Each agent is defined by an `agent.mjs` file with `export const agent`:
@@ -2098,7 +2083,7 @@ node --version
 The core library provides schema validation, API execution, and MCP server activation.
 
 ```bash
-npm install flowmcp-core
+npm install github:FlowMCP/flowmcp-core
 ```
 
 ```javascript
@@ -2224,7 +2209,7 @@ For a new project using FlowMCP, a minimal `package.json` looks like this:
     "version": "1.0.0",
     "type": "module",
     "dependencies": {
-        "flowmcp-core": "latest",
+        "flowmcp-core": "github:FlowMCP/flowmcp-core",
         "@modelcontextprotocol/sdk": "latest"
     }
 }
@@ -2264,7 +2249,7 @@ Understand the architecture and data flow. See How It Works.
    mkdir my-flowmcp-project
    cd my-flowmcp-project
    npm init -y
-   npm install flowmcp-core
+   npm install github:FlowMCP/flowmcp-core
    ```
 
    Add `"type": "module"` to your `package.json` for ES module support.
@@ -2416,16 +2401,6 @@ FlowMCP is a **schema-driven normalization layer** that transforms any data sour
 
 No custom server code. No boilerplate. One schema per provider.
 
-```mermaid
-flowchart LR
-    E[REST API] --> A
-    F[SQLite DB] --> A
-    G[Workflow] --> A
-    A[Schema .mjs] --> B[FlowMCP Runtime]
-    B --> C[MCP Server]
-    C --> D[AI Agent]
-```
-
 ## Four Primitives
 
 FlowMCP v3.0.0 supports four primitives in a single schema file:
@@ -5265,13 +5240,6 @@ export const main = {
 
 The `origin` field determines where the runtime looks for the `.db` file:
 
-```mermaid
-flowchart TD
-    A["origin: 'global'"] --> B["~/.flowmcp/data/companies.db"]
-    C["origin: 'project'"] --> D[".flowmcp/data/companies.db"]
-    E["origin: 'inline'"] --> F["./data/companies.db (relative to schema)"]
-```
-
 | Origin | Path Resolution | Best For |
 |--------|----------------|----------|
 | `global` | `~/.flowmcp/data/{database}` | Shared datasets used across projects |
@@ -5413,16 +5381,6 @@ Skills are instructional multi-step workflows embedded in a schema. Unlike Promp
 
 How a skill is resolved and executed at runtime:
 
-```mermaid
-flowchart LR
-    A[User Input] --> B[Skill Content]
-    B --> C[Replace Placeholders]
-    C --> D[LLM Execution]
-    D --> E["Step 1: Call {{tool:X}}"]
-    E --> F["Step 2: Call {{tool:Y}}"]
-    F --> G[Structured Output]
-```
-
 The skill's `content` is a template. Placeholders like `{{input:tokenSymbol}}` are replaced with user-provided values, and `{{tool:computeRSI}}` references tell the LLM which tool to call at each step.
 
 ## Defining Skills
@@ -5598,19 +5556,6 @@ This page focuses on practical tool creation. See Schema Format for the full spe
 
 A schema file has two parts: the declarative `main` export that describes what tools do, and the optional `handlers` export that transforms requests and responses.
 
-```mermaid
-flowchart TD
-    A[export const main] --> B[namespace + name + root]
-    A --> C[tools]
-    C --> D[method + path]
-    C --> E[parameters]
-    C --> F[description]
-
-    A --> G[export const handlers]
-    G --> H[preRequest]
-    G --> I[postRequest]
-```
-
 ## The `main` Export
 
 The `main` export is a static, JSON-serializable object. No functions, no dynamic values, no imports.
@@ -5903,15 +5848,6 @@ npm install -g flowmcp
 
 The CLI follows a three-step pattern: discover tools, activate them, then call them.
 
-```mermaid
-flowchart LR
-    A[flowmcp search] --> B[Find tools]
-    B --> C[flowmcp add]
-    C --> D[Activate tool]
-    D --> E[flowmcp call]
-    E --> F[Execute tool]
-```
-
 ## Core Commands
 
 | Command | Description |
@@ -6034,18 +5970,6 @@ Instead of writing custom server code for each API, you declare schemas and Flow
 
 How FlowMCP bridges AI clients and APIs:
 
-```mermaid
-flowchart LR
-    A[AI Client] -->|MCP Protocol| B[FlowMCP MCP Server]
-    B --> C[Schema 1: CoinGecko]
-    B --> D[Schema 2: Etherscan]
-    B --> E[Schema 3: DeFi Llama]
-
-    C --> F[CoinGecko API]
-    D --> G[Etherscan API]
-    E --> H[DeFi Llama API]
-```
-
 The AI client sends tool calls over the MCP protocol. FlowMCP resolves the correct schema, validates parameters, calls the upstream API, and returns the result.
 
 ## Starting the Server

src/content/docs/de/docs/agents/overview.md

@@ -13,21 +13,6 @@ Diese Seite behandelt die praktische Anleitung zum Erstellen von Agents. Fuer di
 
 Ein Agent-Manifest (`agent.mjs`) deklariert alles, was der Agent braucht: welche Tools verwendet werden, welches Modell angesteuert wird, wie sich der Agent verhalten soll und wie seine Funktion verifiziert wird.
 
-```mermaid
-flowchart LR
-    A[agent.mjs] --> B[Tool-Auswahl]
-    A --> C[Model Binding]
-    A --> D[System Prompt]
-    A --> E[Prompts]
-    A --> F[Skills]
-    A --> G[Resources]
-    A --> H[Tests]
-
-    B --> I["coingecko-com/tool/simplePrice"]
-    B --> J["etherscan-io/tool/getContractAbi"]
-    B --> K["defillama-com/tool/getProtocolTvl"]
-```
-
 ## Agent-Manifest
 
 Jeder Agent wird durch eine `agent.mjs`-Datei mit `export const agent` definiert:

src/content/docs/de/docs/getting-started/installation.md

@@ -20,7 +20,7 @@ node --version
 Die Core Library bietet Schema-Validierung, API-Ausfuehrung und MCP-Server-Aktivierung.
 
 ```bash
-npm install flowmcp-core
+npm install github:FlowMCP/flowmcp-core
 ```
 
 ```javascript
@@ -146,7 +146,7 @@ Fuer ein neues Projekt mit FlowMCP sieht eine minimale `package.json` so aus:
     "version": "1.0.0",
     "type": "module",
     "dependencies": {
-        "flowmcp-core": "latest",
+        "flowmcp-core": "github:FlowMCP/flowmcp-core",
         "@modelcontextprotocol/sdk": "latest"
     }
 }

src/content/docs/de/docs/getting-started/quickstart.md

@@ -18,7 +18,7 @@ description: "Erstelle dein erstes FlowMCP-Schema und rufe eine API in 5 Minuten
    mkdir my-flowmcp-project
    cd my-flowmcp-project
    npm init -y
-   npm install flowmcp-core
+   npm install github:FlowMCP/flowmcp-core
    ```
 
    Fuege `"type": "module"` zu deiner `package.json` fuer ES-Module-Unterstuetzung hinzu.