mcp-dokploy-fullapi-proxy 1.0.1

package/.github/workflows/publish.yml

@@ -0,0 +1,30 @@
+name: Publish to npm
+on:
+  push:
+    branches:
+      - main
+jobs:
+  publish:
+    runs-on: ubuntu-latest
+    if: "!contains(github.event.head_commit.message, '[skip ci]')"
+    permissions:
+      contents: write
+    steps:
+      - uses: actions/checkout@v4
+        with:
+          token: ${{ secrets.GITHUB_TOKEN }}
+      - uses: actions/setup-node@v4
+        with:
+          node-version: 20
+          registry-url: https://registry.npmjs.org
+      - run: npm ci
+      - run: npm run build
+      - name: Bump version, publish, push back
+        env:
+          NODE_AUTH_TOKEN: ${{ secrets.NPM_TOKEN }}
+        run: |
+          git config user.name "github-actions"
+          git config user.email "github-actions@github.com"
+          npm version patch -m "v%s [skip ci]"
+          npm publish --access public || echo "Version already published, skipping"
+          git push || echo "Push failed, version published to npm"

package/README.md

@@ -0,0 +1,525 @@
+# mcp-dokploy-fullapi-proxy
+
+> **Dokploy Version: `v0.28.1` — 436 Endpoints** _(last updated: 2026-02-28)_
+
+Full API proxy MCP for the entire Dokploy API. **1 tool, ~200 tokens** instead of 67+ tools consuming ~35,000 tokens per conversation.
+
+## How it works
+
+A single MCP tool `dokploy(method, params?)` acts as a thin proxy to Dokploy's tRPC API. All intelligence lives in skill files that Claude reads on-demand via **Progressive Disclosure** – only the relevant API section is loaded into context, saving 77–92% tokens per request.
+
+## Setup
+
+```bash
+git clone https://github.com/hl9020/mcp-dokploy-fullapi-proxy.git
+cd mcp-dokploy-fullapi-proxy
+npm install
+npm run build
+```
+
+## Configuration
+
+> **Architecture:** This MCP uses a two-part setup:
+> 1. **MCP Server** – gives the AI tool access to the `dokploy()` function
+> 2. **Skill/Instructions** – teaches the AI *which* endpoints exist and *how* to call them
+>
+> Without the skill, the AI has the tool but doesn't know the API. Tools that support native skill files get Progressive Disclosure (on-demand loading). Tools without skill support need the SKILL.md content injected as instructions (higher baseline token cost but still far better than 67 individual tools).
+
+---
+
+### Claude Desktop / Claude.ai
+
+**MCP:** ✅ Native  |  **Skills:** ✅ Native (ZIP upload)
+
+#### 1. MCP Server
+
+Add to `claude_desktop_config.json`:
+- **Windows:** `%APPDATA%\Claude\claude_desktop_config.json`
+- **macOS:** `~/Library/Application Support/Claude/claude_desktop_config.json`
+
+```json
+{
+  "mcpServers": {
+    "dokploy-fullapi-proxy": {
+      "command": "node",
+      "args": ["/path/to/mcp-dokploy-fullapi-proxy/dist/index.js"],
+      "env": {
+        "DOKPLOY_URL": "https://your-dokploy-instance.com/api",
+        "DOKPLOY_TOKEN": "your-api-token"
+      }
+    }
+  }
+}
+```
+
+#### 2. Skill Upload
+
+1. Download `skill/dokploy-api.zip` from this repo
+2. Go to **Claude.ai / Claude Desktop → Customize → Skills**
+3. Click **+** and upload `dokploy-api.zip`
+
+
+Get your API token from Dokploy: **Settings → Profile → API/Token Section**.
+
+---
+
+### Claude Code
+
+**MCP:** ✅ Native  |  **Skills:** ✅ Native (`.claude/skills/` directory)
+
+#### 1. MCP Server
+
+```bash
+claude mcp add dokploy-fullapi-proxy \
+  -e DOKPLOY_URL=https://your-dokploy-instance.com/api \
+  -e DOKPLOY_TOKEN=your-api-token \
+  -- node /path/to/mcp-dokploy-fullapi-proxy/dist/index.js
+```
+
+Or add to `.mcp.json` in your project root (Windows: use `cmd /c` wrapper):
+
+```json
+{
+  "mcpServers": {
+    "dokploy-fullapi-proxy": {
+      "command": "node",
+      "args": ["/path/to/mcp-dokploy-fullapi-proxy/dist/index.js"],
+      "env": {
+        "DOKPLOY_URL": "https://your-dokploy-instance.com/api",
+        "DOKPLOY_TOKEN": "your-api-token"
+      }
+    }
+  }
+}
+```
+
+#### 2. Skills
+
+Copy the `skill/` directory into your project's Claude Code skills folder:
+
+```bash
+mkdir -p .claude/skills/dokploy-api
+cp skill/*.md .claude/skills/dokploy-api/
+```
+
+Claude Code uses Progressive Disclosure from `.claude/skills/` automatically – it reads `SKILL.md` first, then loads the relevant reference file on demand.
+
+---
+
+### Cursor
+
+**MCP:** ✅ Native  |  **Skills:** ✅ Agent Skills (auto-discovered)
+
+#### 1. MCP Server
+
+Add to `~/.cursor/mcp.json` (global) or `.cursor/mcp.json` (project):
+
+```json
+{
+  "mcpServers": {
+    "dokploy-fullapi-proxy": {
+      "command": "node",
+      "args": ["/path/to/mcp-dokploy-fullapi-proxy/dist/index.js"],
+      "env": {
+        "DOKPLOY_URL": "https://your-dokploy-instance.com/api",
+        "DOKPLOY_TOKEN": "your-api-token"
+      }
+    }
+  }
+}
+```
+
+#### 2. Skills
+
+Cursor supports **Agent Skills** – an open standard for extending AI agents. Copy the skill files into your project:
+
+```bash
+mkdir -p .cursor/skills/dokploy-api
+cp skill/*.md .cursor/skills/dokploy-api/
+```
+
+Cursor auto-discovers skills and applies them as agent-decided rules. Enable under **Settings → Rules → Import Settings → Agent Skills**.
+
+Alternatively, create a `.cursor/rules/dokploy.mdc` rule:
+
+```markdown
+---
+description: Use when managing Dokploy infrastructure
+alwaysApply: false
+---
+[Paste contents of skill/SKILL.md here]
+```
+
+---
+
+### Windsurf
+
+**MCP:** ✅ Native  |  **Skills:** ❌ No native support (use Rules workaround)
+
+#### 1. MCP Server
+
+Add to `~/.codeium/windsurf/mcp_config.json` (Windows: `%USERPROFILE%\.codeium\windsurf\mcp_config.json`):
+
+```json
+{
+  "mcpServers": {
+    "dokploy-fullapi-proxy": {
+      "command": "node",
+      "args": ["/path/to/mcp-dokploy-fullapi-proxy/dist/index.js"],
+      "env": {
+        "DOKPLOY_URL": "https://your-dokploy-instance.com/api",
+        "DOKPLOY_TOKEN": "your-api-token"
+      }
+    }
+  }
+}
+```
+
+
+#### 2. Skill Workaround (Global Rule)
+
+Create a global or workspace rule via **Settings → Rules, Memories & Workflows → Rules → + Global** (or `+ Workspace`), or create `.windsurf/rules/dokploy.md`:
+
+```markdown
+## Dokploy MCP API Reference
+
+When managing Dokploy infrastructure, use the `dokploy(method, params?)` MCP tool.
+
+[Paste contents of skill/SKILL.md here]
+```
+
+---
+
+### VS Code + GitHub Copilot
+
+**MCP:** ✅ Native  |  **Skills:** ❌ No native support (use Instructions workaround)
+
+#### 1. MCP Server
+
+Add to `.vscode/mcp.json` in your workspace:
+
+```json
+{
+  "servers": {
+    "dokploy-fullapi-proxy": {
+      "command": "node",
+      "args": ["/path/to/mcp-dokploy-fullapi-proxy/dist/index.js"],
+      "env": {
+        "DOKPLOY_URL": "https://your-dokploy-instance.com/api",
+        "DOKPLOY_TOKEN": "your-api-token"
+      }
+    }
+  }
+}
+```
+
+#### 2. Skill Workaround (Copilot Instructions)
+
+Create `.github/copilot-instructions.md` in your project root and paste the contents of `skill/SKILL.md`.
+
+---
+
+### Cline (VS Code Extension)
+
+**MCP:** ✅ Native  |  **Skills:** ❌ No native support (use .clinerules workaround)
+
+#### 1. MCP Server
+
+Open Cline → **MCP Servers** icon → **Edit MCP Settings**, then add:
+
+```json
+{
+  "mcpServers": {
+    "dokploy-fullapi-proxy": {
+      "command": "node",
+      "args": ["/path/to/mcp-dokploy-fullapi-proxy/dist/index.js"],
+      "env": {
+        "DOKPLOY_URL": "https://your-dokploy-instance.com/api",
+        "DOKPLOY_TOKEN": "your-api-token"
+      }
+    }
+  }
+}
+```
+
+#### 2. Skill Workaround (.clinerules)
+
+Create a `.clinerules` file in your project root and paste the contents of `skill/SKILL.md`. Cline injects this into every conversation as custom instructions.
+
+---
+
+### Continue.dev
+
+**MCP:** ✅ Native  |  **Skills:** ❌ No native support (use Rules workaround)
+
+#### 1. MCP Server
+
+Create `.continue/mcpServers/dokploy.json`:
+
+```json
+{
+  "mcpServers": {
+    "dokploy-fullapi-proxy": {
+      "command": "node",
+      "args": ["/path/to/mcp-dokploy-fullapi-proxy/dist/index.js"],
+      "env": {
+        "DOKPLOY_URL": "https://your-dokploy-instance.com/api",
+        "DOKPLOY_TOKEN": "your-api-token"
+      }
+    }
+  }
+}
+```
+
+#### 2. Skill Workaround (Rules)
+
+Add a `.continue/rules/dokploy.md` rule with the contents of `skill/SKILL.md`. Continue applies rules as persistent context for the model.
+
+---
+
+### OpenAI Codex CLI
+
+**MCP:** ✅ Native  |  **Skills:** ✅ Native (`.agents/skills/` directory)
+
+#### 1. MCP Server
+
+```bash
+codex mcp add dokploy-fullapi-proxy \
+  --env DOKPLOY_URL=https://your-dokploy-instance.com/api \
+  --env DOKPLOY_TOKEN=your-api-token \
+  -- node /path/to/mcp-dokploy-fullapi-proxy/dist/index.js
+```
+
+Or add to `~/.codex/config.toml` (global) or `.codex/config.toml` (project):
+
+```toml
+[mcp_servers.dokploy-fullapi-proxy]
+command = "node"
+args = ["/path/to/mcp-dokploy-fullapi-proxy/dist/index.js"]
+
+[mcp_servers.dokploy-fullapi-proxy.env]
+DOKPLOY_URL = "https://your-dokploy-instance.com/api"
+DOKPLOY_TOKEN = "your-api-token"
+```
+
+#### 2. Skills
+
+Copy the `skill/` directory into `.agents/skills/`:
+
+```bash
+mkdir -p .agents/skills/dokploy-api
+cp skill/*.md .agents/skills/dokploy-api/
+```
+
+Codex discovers skills from `.agents/skills/` automatically via `AGENTS.md`.
+
+
+---
+
+### Zed
+
+**MCP:** ✅ Native  |  **Skills:** ❌ No native support (use AGENTS.md workaround)
+
+#### 1. MCP Server
+
+Add to Zed `settings.json` (via **Agent Panel → Settings**):
+
+```json
+{
+  "context_servers": {
+    "dokploy-fullapi-proxy": {
+      "command": {
+        "path": "node",
+        "args": ["/path/to/mcp-dokploy-fullapi-proxy/dist/index.js"],
+        "env": {
+          "DOKPLOY_URL": "https://your-dokploy-instance.com/api",
+          "DOKPLOY_TOKEN": "your-api-token"
+        }
+      }
+    }
+  }
+}
+```
+
+#### 2. Skill Workaround
+
+Create an `AGENTS.md` file in your project root and include the contents of `skill/SKILL.md`. Zed's agent reads `AGENTS.md` for project instructions.
+
+---
+
+### Google Antigravity
+
+**MCP:** ✅ Native  |  **Skills:** ❌ No native support (use Rules workaround)
+
+#### 1. MCP Server
+
+Add to `~/.gemini/settings.json` or via **Antigravity Settings → MCP**:
+
+```json
+{
+  "mcpServers": {
+    "dokploy-fullapi-proxy": {
+      "command": "node",
+      "args": ["/path/to/mcp-dokploy-fullapi-proxy/dist/index.js"],
+      "env": {
+        "DOKPLOY_URL": "https://your-dokploy-instance.com/api",
+        "DOKPLOY_TOKEN": "your-api-token"
+      }
+    }
+  }
+}
+```
+
+#### 2. Skill Workaround
+
+Create a `GEMINI.md` or project rules file and include the contents of `skill/SKILL.md`.
+
+---
+
+### Roo Code (VS Code Extension)
+
+**MCP:** ✅ Native  |  **Skills:** ❌ No native support (use Custom Instructions workaround)
+
+#### 1. MCP Server
+
+Configure via Roo Code's MCP settings panel, same JSON format as Cline:
+
+```json
+{
+  "mcpServers": {
+    "dokploy-fullapi-proxy": {
+      "command": "node",
+      "args": ["/path/to/mcp-dokploy-fullapi-proxy/dist/index.js"],
+      "env": {
+        "DOKPLOY_URL": "https://your-dokploy-instance.com/api",
+        "DOKPLOY_TOKEN": "your-api-token"
+      }
+    }
+  }
+}
+```
+
+#### 2. Skill Workaround
+
+Add the contents of `skill/SKILL.md` to Roo Code's custom system prompt or `.roo/rules/` directory.
+
+---
+
+## Compatibility Matrix
+
+| Tool | MCP Server | Native Skills | Skill Workaround | Config Location |
+|------|:---:|:---:|:---:|---|
+| **Claude Desktop / Claude.ai** | ✅ | ✅ ZIP Upload | – | `claude_desktop_config.json` |
+| **Claude Code** | ✅ | ✅ `.claude/skills/` | – | `.mcp.json` |
+| **Cursor** | ✅ | ✅ Agent Skills | `.cursor/rules/*.mdc` | `.cursor/mcp.json` |
+| **Codex CLI** | ✅ | ✅ `.agents/skills/` | – | `.codex/config.toml` |
+| **Windsurf** | ✅ | ❌ | `.windsurf/rules/` | `~/.codeium/windsurf/mcp_config.json` |
+| **VS Code + Copilot** | ✅ | ❌ | `.github/copilot-instructions.md` | `.vscode/mcp.json` |
+| **Cline** | ✅ | ❌ | `.clinerules` | MCP Settings JSON |
+| **Continue.dev** | ✅ | ❌ | `.continue/rules/` | `.continue/mcpServers/*.json` |
+| **Zed** | ✅ | ❌ | `AGENTS.md` | `settings.json` |
+| **Google Antigravity** | ✅ | ❌ | `GEMINI.md` / Rules | `~/.gemini/settings.json` |
+| **Roo Code** | ✅ | ❌ | `.roo/rules/` | MCP Settings JSON |
+
+> **Legend:** Tools with native skill support get Progressive Disclosure (on-demand file loading, ~500–2,000 tokens per request). Tools without skills load the full SKILL.md as instructions (~500 tokens always-on, without the granular per-resource files).
+
+---
+
+## Skill files
+
+The `skill/` directory contains API documentation split by resource (auto-generated from Dokploy's OpenAPI spec). Claude reads only the relevant section when needed:
+
+| File | Endpoints | Coverage |
+|------|-----------|----------|
+| `SKILL.md` | – | Entry point, routing table |
+| `project.md` | 12 | Projects & Environments |
+| `app.md` | 28 | Applications |
+| `compose.md` | 27 | Compose services |
+| `domain.md` | 9 | Domains & SSL |
+| `database.md` | 65 | PostgreSQL, MySQL, MariaDB, MongoDB, Redis |
+| `deployment.md` | 12 | Deployments, Preview, Rollback |
+| `docker.md` | 7 | Docker containers |
+| `server.md` | 23 | Server, Cluster, Swarm |
+| `notification.md` | 38 | Notifications (Slack, Discord, Telegram, Email, Teams, Resend, ...) |
+| `settings.md` | 73 | Settings, Admin, Stripe, SSO, LicenseKey |
+| `user.md` | 27 | User & Organization |
+| `git.md` | 30 | Git Providers (GitHub, GitLab, Bitbucket, Gitea) |
+| `infra.md` | 85 | Mounts, Redirects, Security, Ports, Backups, Schedule, Certs, Registry, SSH, AI |
+
+## Token comparison
+
+| | Official Dokploy MCP | mcp-dokploy-fullapi-proxy |
+|---|---|---|
+| Tools registered | 67 | 1 |
+| Permanent context tokens | ~35,000 | ~200 |
+| API coverage | ~16% (67 of 436 endpoints) | 100% (436 endpoints) |
+| On-demand tokens per request | 0 | ~500–2,000 (1 skill file) |
+| Typical savings | – | **77–92% fewer tokens** |
+
+
+## How the skill system works
+
+```
+User: "Deploy my app"
+  ↓
+Claude reads SKILL.md routing table (~500 tokens)
+  ↓
+Claude reads app.md (~1,600 tokens)
+  ↓
+Claude calls: dokploy("application.deploy", { applicationId: "..." })
+```
+
+Instead of loading all 407 endpoint definitions (~15,500 tokens) into every conversation, Claude loads only what's needed. A typical request costs ~2,100 tokens instead of ~15,500.
+
+## Building the skill ZIP
+
+```bash
+# Windows (PowerShell)
+Compress-Archive -Path skill\* -DestinationPath dokploy-api.zip -Force
+
+# macOS / Linux
+cd skill && zip -r ../dokploy-api.zip . && cd ..
+```
+
+## Environment variables
+
+| Variable | Required | Default | Description |
+|----------|----------|---------|-------------|
+| `DOKPLOY_URL` | No | `http://localhost:3000/api` | Dokploy API base URL |
+| `DOKPLOY_TOKEN` | Yes | – | API authentication token |
+
+Get your API token from Dokploy: **Settings → Profile → API/Token Section**.
+
+## pick - Response Filter
+
+The `pick` parameter filters large API responses clientside to only the fields you need — drastically reducing token usage on endpoints like `project.all`.
+
+```js
+// Without pick: returns entire project tree (~50KB, ~10,000 tokens)
+dokploy("project.all")
+
+// With pick: returns only MySQL instances (~200 tokens)
+dokploy("project.all", {}, pick: ["mysqlId", "name", "appName", "applicationStatus"])
+
+// Get backup info without noise
+dokploy("mysql.one", { mysqlId: "..." }, pick: ["backupId", "schedule", "enabled", "database"])
+```
+
+**How it works:** Recursively traverses the JSON response and retains only nodes that contain the specified field names. Empty objects/arrays are removed automatically.
+
+**When to use it:**
+- `project.all` → always use `pick` when looking for a specific resource type
+- `*.one` endpoints → when only a sub-section (backups, mounts, etc.) is needed
+- Any endpoint returning large nested objects
+
+## Verify
+
+Start a new conversation and ask:
+```
+Show me all Dokploy projects
+```
+The AI should call `dokploy("project.all")`.
+
+## License
+
+MIT

package/dist/index.d.ts

@@ -0,0 +1 @@
+export {};

package/dist/index.js

@@ -0,0 +1,105 @@
+import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js";
+import { StdioServerTransport } from "@modelcontextprotocol/sdk/server/stdio.js";
+import { z } from "zod";
+const DOKPLOY_URL = process.env.DOKPLOY_URL ?? "http://localhost:3000/api";
+const DOKPLOY_TOKEN = process.env.DOKPLOY_TOKEN ?? process.env.DOKPLOY_API_KEY ?? "";
+if (!DOKPLOY_TOKEN) {
+    console.error("DOKPLOY_TOKEN env var required");
+    process.exit(1);
+}
+const server = new McpServer({
+    name: "mcp-dokploy-fullapi-proxy",
+    version: "1.0.0",
+});
+function pickFields(data, fields) {
+    if (!data || typeof data !== "object")
+        return data;
+    if (Array.isArray(data))
+        return data.map(item => pickFields(item, fields));
+    const obj = data;
+    const result = {};
+    for (const key of Object.keys(obj)) {
+        if (fields.includes(key)) {
+            result[key] = obj[key];
+        }
+        else if (typeof obj[key] === "object" && obj[key] !== null) {
+            const nested = pickFields(obj[key], fields);
+            // nur hinzufügen wenn nested nicht leer ist
+            if (Array.isArray(nested) ? nested.length > 0 : Object.keys(nested).length > 0) {
+                result[key] = nested;
+            }
+        }
+    }
+    return result;
+}
+server.tool("dokploy", "Universal Dokploy API proxy. Calls any tRPC endpoint. " +
+    "Use method like 'project.all', 'application.deploy', 'compose.update'. " +
+    "Params are passed as JSON body to the API. " +
+    "Optional 'pick' filters the response to only include specified field names (e.g. pick: ['mysql'] on project.all returns only mysql arrays).", { method: z.string(), params: z.record(z.string(), z.unknown()).optional(), pick: z.array(z.string()).optional() }, async ({ method, params, pick }) => {
+    const baseUrl = `${DOKPLOY_URL}/trpc/${method}`;
+    const hasParams = params && Object.keys(params).length > 0;
+    try {
+        let res;
+        if (hasParams) {
+            // tRPC GET queries: params as ?input=<JSON>
+            const qs = new URLSearchParams({ input: JSON.stringify({ json: params }) });
+            res = await fetch(`${baseUrl}?${qs}`, {
+                method: "GET",
+                headers: { "x-api-key": DOKPLOY_TOKEN },
+            });
+            // Fallback to POST for mutations
+            if (res.status === 405) {
+                res = await fetch(baseUrl, {
+                    method: "POST",
+                    headers: { "Content-Type": "application/json", "x-api-key": DOKPLOY_TOKEN },
+                    body: JSON.stringify({ json: params }),
+                });
+            }
+        }
+        else {
+            res = await fetch(baseUrl, {
+                method: "GET",
+                headers: { "x-api-key": DOKPLOY_TOKEN },
+            });
+            if (res.status === 405) {
+                res = await fetch(baseUrl, {
+                    method: "POST",
+                    headers: { "Content-Type": "application/json", "x-api-key": DOKPLOY_TOKEN },
+                    body: "{}",
+                });
+            }
+        }
+        const text = await res.text();
+        let data;
+        try {
+            data = JSON.parse(text);
+        }
+        catch {
+            data = text;
+        }
+        if (!res.ok) {
+            const msg = typeof data === "object" ? JSON.stringify(data, null, 2) : String(data);
+            return { content: [{ type: "text", text: `❌ ${res.status} ${res.statusText}\n${msg}` }], isError: true };
+        }
+        const result = typeof data === "object" && data !== null && "result" in data
+            ? data.result
+            : data;
+        const output = typeof result === "object" && result !== null && "data" in result
+            ? result.data
+            : result;
+        const filtered = pick && pick.length > 0 ? pickFields(output, pick) : output;
+        return { content: [{ type: "text", text: JSON.stringify(filtered, null, 2) }] };
+    }
+    catch (e) {
+        return { content: [{ type: "text", text: `❌ Network error: ${e instanceof Error ? e.message : String(e)}` }], isError: true };
+    }
+});
+async function main() {
+    const transport = new StdioServerTransport();
+    await server.connect(transport);
+}
+main().catch((e) => {
+    console.error("Fatal:", e);
+    process.exit(1);
+});
+//# sourceMappingURL=index.js.map

package/dist/index.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"index.js","sourceRoot":"","sources":["../src/index.ts"],"names":[],"mappings":"AAAA,OAAO,EAAE,SAAS,EAAE,MAAM,yCAAyC,CAAC;AACpE,OAAO,EAAE,oBAAoB,EAAE,MAAM,2CAA2C,CAAC;AACjF,OAAO,EAAE,CAAC,EAAE,MAAM,KAAK,CAAC;AAExB,MAAM,WAAW,GAAG,OAAO,CAAC,GAAG,CAAC,WAAW,IAAI,2BAA2B,CAAC;AAC3E,MAAM,aAAa,GAAG,OAAO,CAAC,GAAG,CAAC,aAAa,IAAI,OAAO,CAAC,GAAG,CAAC,eAAe,IAAI,EAAE,CAAC;AAErF,IAAI,CAAC,aAAa,EAAE,CAAC;IACnB,OAAO,CAAC,KAAK,CAAC,gCAAgC,CAAC,CAAC;IAChD,OAAO,CAAC,IAAI,CAAC,CAAC,CAAC,CAAC;AAClB,CAAC;AAED,MAAM,MAAM,GAAG,IAAI,SAAS,CAAC;IAC3B,IAAI,EAAE,2BAA2B;IACjC,OAAO,EAAE,OAAO;CACjB,CAAC,CAAC;AAEH,SAAS,UAAU,CAAC,IAAa,EAAE,MAAgB;IACjD,IAAI,CAAC,IAAI,IAAI,OAAO,IAAI,KAAK,QAAQ;QAAE,OAAO,IAAI,CAAC;IACnD,IAAI,KAAK,CAAC,OAAO,CAAC,IAAI,CAAC;QAAE,OAAO,IAAI,CAAC,GAAG,CAAC,IAAI,CAAC,EAAE,CAAC,UAAU,CAAC,IAAI,EAAE,MAAM,CAAC,CAAC,CAAC;IAC3E,MAAM,GAAG,GAAG,IAA+B,CAAC;IAC5C,MAAM,MAAM,GAA4B,EAAE,CAAC;IAC3C,KAAK,MAAM,GAAG,IAAI,MAAM,CAAC,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC;QACnC,IAAI,MAAM,CAAC,QAAQ,CAAC,GAAG,CAAC,EAAE,CAAC;YACzB,MAAM,CAAC,GAAG,CAAC,GAAG,GAAG,CAAC,GAAG,CAAC,CAAC;QACzB,CAAC;aAAM,IAAI,OAAO,GAAG,CAAC,GAAG,CAAC,KAAK,QAAQ,IAAI,GAAG,CAAC,GAAG,CAAC,KAAK,IAAI,EAAE,CAAC;YAC7D,MAAM,MAAM,GAAG,UAAU,CAAC,GAAG,CAAC,GAAG,CAAC,EAAE,MAAM,CAAC,CAAC;YAC5C,4CAA4C;YAC5C,IAAI,KAAK,CAAC,OAAO,CAAC,MAAM,CAAC,CAAC,CAAC,CAAE,MAAoB,CAAC,MAAM,GAAG,CAAC,CAAC,CAAC,CAAC,MAAM,CAAC,IAAI,CAAC,MAAgB,CAAC,CAAC,MAAM,GAAG,CAAC,EAAE,CAAC;gBACxG,MAAM,CAAC,GAAG,CAAC,GAAG,MAAM,CAAC;YACvB,CAAC;QACH,CAAC;IACH,CAAC;IACD,OAAO,MAAM,CAAC;AAChB,CAAC;AAED,MAAM,CAAC,IAAI,CACT,SAAS,EACT,wDAAwD;IACtD,yEAAyE;IACzE,6CAA6C;IAC7C,6IAA6I,EAC/I,EAAE,MAAM,EAAE,CAAC,CAAC,MAAM,EAAE,EAAE,MAAM,EAAE,CAAC,CAAC,MAAM,CAAC,CAAC,CAAC,MAAM,EAAE,EAAE,CAAC,CAAC,OAAO,EAAE,CAAC,CAAC,QAAQ,EAAE,EAAE,IAAI,EAAE,CAAC,CAAC,KAAK,CAAC,CAAC,CAAC,MAAM,EAAE,CAAC,CAAC,QAAQ,EAAE,EAAE,EAClH,KAAK,EAAE,EAAE,MAAM,EAAE,MAAM,EAAE,IAAI,EAAE,EAAE,EAAE;IACjC,MAAM,OAAO,GAAG,GAAG,WAAW,SAAS,MAAM,EAAE,CAAC;IAChD,MAAM,SAAS,GAAG,MAAM,IAAI,MAAM,CAAC,IAAI,CAAC,MAAM,CAAC,CAAC,MAAM,GAAG,CAAC,CAAC;IAE3D,IAAI,CAAC;QACH,IAAI,GAAa,CAAC;QAClB,IAAI,SAAS,EAAE,CAAC;YACd,4CAA4C;YAC5C,MAAM,EAAE,GAAG,IAAI,eAAe,CAAC,EAAE,KAAK,EAAE,IAAI,CAAC,SAAS,CAAC,EAAE,IAAI,EAAE,MAAM,EAAE,CAAC,EAAE,CAAC,CAAC;YAC5E,GAAG,GAAG,MAAM,KAAK,CAAC,GAAG,OAAO,IAAI,EAAE,EAAE,EAAE;gBACpC,MAAM,EAAE,KAAK;gBACb,OAAO,EAAE,EAAE,WAAW,EAAE,aAAa,EAAE;aACxC,CAAC,CAAC;YACH,iCAAiC;YACjC,IAAI,GAAG,CAAC,MAAM,KAAK,GAAG,EAAE,CAAC;gBACvB,GAAG,GAAG,MAAM,KAAK,CAAC,OAAO,EAAE;oBACzB,MAAM,EAAE,MAAM;oBACd,OAAO,EAAE,EAAE,cAAc,EAAE,kBAAkB,EAAE,WAAW,EAAE,aAAa,EAAE;oBAC3E,IAAI,EAAE,IAAI,CAAC,SAAS,CAAC,EAAE,IAAI,EAAE,MAAM,EAAE,CAAC;iBACvC,CAAC,CAAC;YACL,CAAC;QACH,CAAC;aAAM,CAAC;YACN,GAAG,GAAG,MAAM,KAAK,CAAC,OAAO,EAAE;gBACzB,MAAM,EAAE,KAAK;gBACb,OAAO,EAAE,EAAE,WAAW,EAAE,aAAa,EAAE;aACxC,CAAC,CAAC;YACH,IAAI,GAAG,CAAC,MAAM,KAAK,GAAG,EAAE,CAAC;gBACvB,GAAG,GAAG,MAAM,KAAK,CAAC,OAAO,EAAE;oBACzB,MAAM,EAAE,MAAM;oBACd,OAAO,EAAE,EAAE,cAAc,EAAE,kBAAkB,EAAE,WAAW,EAAE,aAAa,EAAE;oBAC3E,IAAI,EAAE,IAAI;iBACX,CAAC,CAAC;YACL,CAAC;QACH,CAAC;QAED,MAAM,IAAI,GAAG,MAAM,GAAG,CAAC,IAAI,EAAE,CAAC;QAC9B,IAAI,IAAa,CAAC;QAClB,IAAI,CAAC;YAAC,IAAI,GAAG,IAAI,CAAC,KAAK,CAAC,IAAI,CAAC,CAAC;QAAC,CAAC;QAAC,MAAM,CAAC;YAAC,IAAI,GAAG,IAAI,CAAC;QAAC,CAAC;QAEvD,IAAI,CAAC,GAAG,CAAC,EAAE,EAAE,CAAC;YACZ,MAAM,GAAG,GAAG,OAAO,IAAI,KAAK,QAAQ,CAAC,CAAC,CAAC,IAAI,CAAC,SAAS,CAAC,IAAI,EAAE,IAAI,EAAE,CAAC,CAAC,CAAC,CAAC,CAAC,MAAM,CAAC,IAAI,CAAC,CAAC;YACpF,OAAO,EAAE,OAAO,EAAE,CAAC,EAAE,IAAI,EAAE,MAAe,EAAE,IAAI,EAAE,KAAK,GAAG,CAAC,MAAM,IAAI,GAAG,CAAC,UAAU,KAAK,GAAG,EAAE,EAAE,CAAC,EAAE,OAAO,EAAE,IAAI,EAAE,CAAC;QACpH,CAAC;QAED,MAAM,MAAM,GAAG,OAAO,IAAI,KAAK,QAAQ,IAAI,IAAI,KAAK,IAAI,IAAI,QAAQ,IAAI,IAAI;YAC1E,CAAC,CAAE,IAAgC,CAAC,MAAM;YAC1C,CAAC,CAAC,IAAI,CAAC;QACT,MAAM,MAAM,GAAG,OAAO,MAAM,KAAK,QAAQ,IAAI,MAAM,KAAK,IAAI,IAAI,MAAM,IAAI,MAAM;YAC9E,CAAC,CAAE,MAAkC,CAAC,IAAI;YAC1C,CAAC,CAAC,MAAM,CAAC;QAEX,MAAM,QAAQ,GAAG,IAAI,IAAI,IAAI,CAAC,MAAM,GAAG,CAAC,CAAC,CAAC,CAAC,UAAU,CAAC,MAAM,EAAE,IAAI,CAAC,CAAC,CAAC,CAAC,MAAM,CAAC;QAC7E,OAAO,EAAE,OAAO,EAAE,CAAC,EAAE,IAAI,EAAE,MAAe,EAAE,IAAI,EAAE,IAAI,CAAC,SAAS,CAAC,QAAQ,EAAE,IAAI,EAAE,CAAC,CAAC,EAAE,CAAC,EAAE,CAAC;IAC3F,CAAC;IAAC,OAAO,CAAC,EAAE,CAAC;QACX,OAAO,EAAE,OAAO,EAAE,CAAC,EAAE,IAAI,EAAE,MAAe,EAAE,IAAI,EAAE,oBAAoB,CAAC,YAAY,KAAK,CAAC,CAAC,CAAC,CAAC,CAAC,OAAO,CAAC,CAAC,CAAC,MAAM,CAAC,CAAC,CAAC,EAAE,EAAE,CAAC,EAAE,OAAO,EAAE,IAAI,EAAE,CAAC;IACzI,CAAC;AACH,CAAC,CACF,CAAC;AAEF,KAAK,UAAU,IAAI;IACjB,MAAM,SAAS,GAAG,IAAI,oBAAoB,EAAE,CAAC;IAC7C,MAAM,MAAM,CAAC,OAAO,CAAC,SAAS,CAAC,CAAC;AAClC,CAAC;AAED,IAAI,EAAE,CAAC,KAAK,CAAC,CAAC,CAAC,EAAE,EAAE;IACjB,OAAO,CAAC,KAAK,CAAC,QAAQ,EAAE,CAAC,CAAC,CAAC;IAC3B,OAAO,CAAC,IAAI,CAAC,CAAC,CAAC,CAAC;AAClB,CAAC,CAAC,CAAC"}

package/package.json

@@ -0,0 +1,28 @@
+{
+  "name": "mcp-dokploy-fullapi-proxy",
+  "version": "1.0.1",
+  "description": "Full API proxy MCP for the entire Dokploy API – single tool, skill-driven, 407 endpoints",
+  "type": "module",
+  "main": "dist/index.js",
+  "scripts": {
+    "build": "tsc",
+    "dev": "tsx src/index.ts",
+    "start": "node dist/index.js",
+    "update-skills": "python scripts/update-skills.py"
+  },
+  "keywords": [
+    "mcp",
+    "dokploy",
+    "proxy"
+  ],
+  "license": "MIT",
+  "dependencies": {
+    "@modelcontextprotocol/sdk": "^1.12.0",
+    "zod": "^4.3.6"
+  },
+  "devDependencies": {
+    "@types/node": "^22.0.0",
+    "tsx": "^4.19.0",
+    "typescript": "^5.7.0"
+  }
+}