sipeed · 2023478 · Apr 11, 2026 · Apr 25, 2026 · Apr 25, 2026 · Apr 25, 2026
diff --git a/docs/channels/dingtalk.md b/docs/channels/dingtalk.md
@@ -35,19 +35,40 @@ In **Permissions & Scopes**, ensure the following permissions are granted:
 
 ### 5. Configure PicoClaw
 
+#### 1. WebUI Configuration
+
+We recommend using the WebUI first because it is faster and more convenient.
+
+![WebUI DingTalk Connection Interface](/img/channels/webui_dingtalk.png)
+
+Fill in the Client ID (`YOUR_CLIENT_ID`) and Client Secret (`YOUR_CLIENT_SECRET`) in order, then click **Save**.
+
+#### 2. Configuration Files
+
+Edit `~/.picoclaw/.security.yml`:
+
+```yaml
+dingtalk:
+  settings:
+    client_secret: YOUR_CLIENT_SECRET
+```
+
+Edit `~/.picoclaw/config.json`:
+
 ```json
 {
   "channels": {
-    "dingtalk": {
       "enabled": true,
-      "client_id": "YOUR_CLIENT_ID",
-      "client_secret": "YOUR_CLIENT_SECRET",
-      "allow_from": [],
-      "group_trigger": {
-        "mention_only": true
+      "type": "dingtalk",
+      "reasoning_channel_id": "",
+      "group_trigger": {},
+      "typing": {},
+      "placeholder": {
+        "enabled": false
       },
-      "reasoning_channel_id": ""
-    }
+      "settings": {
+        "client_id": "YOUR_CLIENT_ID"
+      }
   }
 }
 ```

diff --git a/docs/channels/discord.md b/docs/channels/discord.md
@@ -24,14 +24,43 @@ title: Discord
 
 ### 4. Configure
 
+#### 1. WebUI Configuration
+
+We recommend using the WebUI first because it is faster and more convenient.
+
+![WebUI Discord Connection Interface](/img/channels/webui_discord.png)
+
+Fill in the Bot Token (`YOUR_BOT_TOKEN`) and Allowed Sources (`YOUR_USER_ID`) in order, then click **Save**.
+
+#### 2. Configuration Files
+
+Edit `~/.picoclaw/.security.yml`:
+
+```yaml
+discord:
+  settings:
+    token: YOUR_BOT_TOKEN
+```
+
+Edit `~/.picoclaw/config.json`:
+
 ```json
 {
   "channels": {
     "discord": {
       "enabled": true,
-      "token": "YOUR_BOT_TOKEN",
-      "allow_from": ["YOUR_USER_ID"],
-      "group_trigger": {
+      "type": "discord",
+      "allow_from": [
+        "YOUR_USER_ID"
+      ],
+      "reasoning_channel_id": "",
+      "group_trigger": {},
+      "typing": {},
+      "placeholder": {
+        "enabled": false
+      },
+      "settings": {
+        "proxy": "",
         "mention_only": false
       }
     }

diff --git a/docs/channels/feishu.md b/docs/channels/feishu.md
@@ -51,30 +51,53 @@ For development/testing, you can leave `encrypt_key` and `verification_token` em
 
 ### 5. Configure PicoClaw
 
+#### 1. WebUI Configuration
+
+We recommend using the WebUI first because it is faster and more convenient.
+
+![WebUI Feishu Connection Interface](/img/channels/webui_feishu.png)
+
+Fill in the App ID (`YOUR_APP_ID`), App Secret (`YOUR_APP_SECRET`), Encrypt Key (`YOUR_ENCRYPT_KEY`), and Verification Token (`YOUR_VERIFICATION_TOKEN`) in order, then click **Save**.
+
+#### 2. Configuration Files
+
+Edit `~/.picoclaw/config.json`:
+
 ```json
 {
   "channels": {
     "feishu": {
       "enabled": true,
-      "app_id": "cli_xxx",
-      "app_secret": "YOUR_APP_SECRET",
-      "encrypt_key": "YOUR_ENCRYPT_KEY",
-      "verification_token": "YOUR_VERIFICATION_TOKEN",
-      "allow_from": [],
-      "group_trigger": {
-        "mention_only": true
-      },
+      "type": "feishu",
+      "allow_from": [
+        "YOUR_USER_ID"
+      ],
+      "reasoning_channel_id": "",
+      "group_trigger": {},
+      "typing": {},
       "placeholder": {
-        "enabled": true,
-        "text": "Thinking..."
+        "enabled": false
       },
-      "random_reaction_emoji": [],
-      "reasoning_channel_id": ""
+      "settings": {
+        "app_id": "YOUR_APP_ID",
+        "random_reaction_emoji": null,
+        "is_lark": false
+      }
     }
   }
 }
 ```
 
+Edit `~/.picoclaw/.security.yml`:
+
+```yaml
+feishu:
+  settings:
+    app_secret: "YOUR_APP_SECRET"
+    encrypt_key: "YOUR_ENCRYPT_KEY"
+    verification_token: "YOUR_VERIFICATION_TOKEN"
+```
+
 ### 6. Publish the App
 
 1. Go to **Version Management & Release**

diff --git a/docs/channels/telegram.md b/docs/channels/telegram.md
@@ -22,13 +22,36 @@ Telegram is the **recommended** channel. It's easy to set up and supports voice
 
 ### 3. Configure
 
+#### 1. WebUI Configuration
+
+We recommend using the WebUI first because it is faster and more convenient.
+
+![WebUI Telegram Connection Interface](/img/channels/webui_telegram.png)
+
+Fill in the Bot Token (`YOUR_BOT_TOKEN`) and Allowed Sources (`YOUR_USER_ID`) in order, then click **Save**.
+
+#### 2. Configuration Files
+
+Edit `~/.picoclaw/.security.yml`:
+
+```yaml
+telegram:
+  settings:
+    token: YOUR_BOT_TOKEN
+```
+
+Edit `~/.picoclaw/config.json`:
+
 ```json
 {
   "channels": {
     "telegram": {
       "enabled": true,
-      "token": "YOUR_BOT_TOKEN",
-      "allow_from": ["YOUR_USER_ID"]
+      "allow_from": [
+        "YOUR_USER_ID"
+      ],
+      "reasoning_channel_id": "",
+      "group_trigger": {}
     }
   }
 }

diff --git a/docs/providers/anthropic.md b/docs/providers/anthropic.md
@@ -0,0 +1,90 @@
+---
+id: Anthropic-api
+title: Claude(Anthropic) API
+---
+
+# Claude (Anthropic) API Configuration Guide
+
+## Overview
+
+**Claude (Anthropic)** offers a **Pay‑as‑you‑go** pricing model. Access is via official Anthropic API Key, with charges based on actual Token consumption.
+
+> ⚠️ **Important Notes**
+>
+> 
+>
+> - Claude **subscription plans** (Pro / Max / Team / Enterprise) and the **API** are **completely separate** products. Subscriptions work only on the claude.ai web, desktop, and mobile apps; they **do NOT include API access** and cannot be used in dev tools like PicoClaw.
+> - The native Anthropic API uses the `/v1/messages` interface format, **not** the OpenAI‑compatible interface (`/v1/chat/completions`). Using OpenAI compatibility mode disables core features such as Prompt Caching, Extended Thinking, and PDF processing.
+> - Using a subscription account’s OAuth Token to call Claude in third‑party tools **violates Anthropic ToS**; this method was officially blocked in April 2026 — do not attempt.
+
+Official docs: https://docs.anthropic.com/en/api/getting-started
+
+
+
+![image-20260409215655752](/img/providers/Claude(Anthropic)1.png)
+
+## Get API Key
+
+1. Visit the Anthropic Console [Anthropic Console](https://console.anthropic.com/), register and log in
+* Go to the API Keys page in the left navigation, click "Create Key" to generate an API Key
+* On the Billing page, add a payment method and top up, API calls will be deducted from the balance based on actual token usage
+
+![image-20260409215733096](/img/providers/Claude(Anthropic)2.png)
+
+### Plan Usage
+
+Pay‑as‑you‑go, charged by actual input/output Tokens consumed. No monthly fee; no charge when idle. Reference rates (prices subject to official changes):
+
+|       Model       | Input (per million Tokens) | Output (per million Tokens) |
+| :---------------: | :------------------------: | :-------------------------: |
+| claude-sonnet-4.6 |             $3             |             $15             |
+|  claude-opus-4.6  |            $15             |             $75             |
+| claude-haiku-4.6  |            $0.8            |             $4              |
+Latest prices follow Anthropic’s official pricing page.
+
+### Supported Models
+
+|           Model            |                         Description                          |
+| :------------------------: | :----------------------------------------------------------: |
+|     claude-sonnet-4.6      | Recommended; balanced performance and cost, excellent long context |
+|      claude-opus-4.6       |    High‑end model, complex reasoning, ultra‑long context     |
+|      claude-haiku-4.6      |             Lightweight, fast response, low cost             |
+| claude-3-5-sonnet-20241022 |          Classic stable version, wide compatibility          |
+> For daily coding, prefer **claude-sonnet-4.6**; switch to Opus only for complex tasks to avoid excessive usage.
+## Configure PicoClaw
+
+### Web UI Configuration
+
+Open PicoClaw WebUI, go to the **Models** page in the left sidebar, and click **Add Model** in the top right corner.
+
+![image-20260409221239555](/img/providers/Claude(Anthropic)3.png)
+|    Field     |                     Value to enter                     |
+| :----------: | :----------------------------------------------------: |
+| Model Alias  |              Custom name, e.g. anthropic               |
+|   Model ID   | anthropic/claude-sonnet-4.6 (or other supported model) |
+|   API Key    |         API Key generated in Anthropic Console         |
+| API Base URL |              https://api.anthropic.com/v1              |
+### Edit Config Files
+config.json 
+```
+    {
+      "model_name": "claude-sonnet-4.6",
+      "model": "anthropic/claude-sonnet-4.6",
+      "api_base": "https://api.anthropic.com/v1"
+    },
+```
+***
+`~/.picoclaw/.security.yml`：
+
+```YAML
+model_list:
+  claude-sonnet-4.6:0:
+    api_keys:
+      - "your-anthropic-api-key"
+```
+## Notes
+
+- Pay‑as‑you‑go: costs scale with usage. Frequent or long‑context tasks can increase costs quickly — set usage alerts on the Console Billing page.
+- The Claude Opus series is significantly more expensive than Sonnet; avoid prolonged use unless necessary.
+- Subscription plans and API balances are separate and non‑transferable: balance added to the Console is for API calls only and unrelated to claude.ai subscriptions.
+
diff --git a/docs/providers/groq.md b/docs/providers/groq.md
@@ -0,0 +1,99 @@
+---
+id: groq-api
+title: Groq API
+---
+# Groq API Configuration Guide
+
+## Overview
+
+Groq is an AI inference platform focused on **ultra-low latency and lightning-fast inference**. Powered by its proprietary Tensor architecture, it delivers exceptional speed in code generation and chat interaction. Supports popular open models like Llama and Mixtral, available via pay-as-you-go.
+
+Official Docs: https://console.groq.com/docs/quickstart
+
+---
+
+## Get API Key
+
+### Step 1: Access Platform
+
+Go to [GroqCloud](https://console.groq.com/home)
+
+register and log in.
+
+### Step 2: Create API Key
+
+1. Go to the API Keys page
+2. Click **Create API Key**
+3. Copy and save your API Key
+
+> ⚠️ **Note**: API Key is shown only once — save it immediately and keep it secure.。
+
+---
+
+![image-20260410163359795](/img/providers/groq1.png)
+
+## Configure PicoClaw
+
+### Pay-as-you-go (Groq only supports pay-as-you-go)
+
+#### Supported Models
+
+|          Model          |                  Description                  |
+| :---------------------: | :-------------------------------------------: |
+| llama-3.3-70b-versatile | Recommended, balanced performance, ultra-fast |
+|    llama-3.3-8b-chat    |     Lightweight, low-cost, fast response     |
+|  mixtral-8x7b-instruct  | Mixture-of-experts, strong long-text handling |
+|      gemma-2-9b-it      |         Google lightweight chat model         |
+
+#### Method 1: Web UI (Recommended)
+
+Open PicoClaw WebUI → go to **Models** → click **Add Model** in the top-right corner.
+
+![image-20260410163514016](/img/providers/groq2.png)
+
+|      Field      |                         Fill Content                         |
+| :--------------: | :----------------------------------------------------------: |
+|   Model Alias   |                  Custom name, e.g.,`groq`                  |
+| Model Identifier | `groq/llama-3.3-70b-versatile` (or other supported models) |
+|     API Key     |                      Your Groq API Key                      |
+|   API Base URL   |              `https://api.groq.com/openai/v1`              |
+
+#### Method 2: Edit Config Files
+
+`config.json`：
+
+```
+{
+  "version": 2,
+  "model_list": [
+    {
+      "model_name": "llama-3.3-70b",
+      "model": "groq/llama-3.3-70b-versatile",
+      "api_base": "https://api.groq.com/openai/v1"
+    },
+  ],
+  "agents": {
+    "defaults": {
+      "model_name": "llama-3.3-70b"
+    }
+  }
+}
+```
+
+`~/.picoclaw/.security.yml`：
+
+```
+model_list:
+  llama-3.3-70b:0:
+    api_keys:
+      - "your-groq-api-key"
+```
+
+---
+
+## Notes
+
+- Groq has **no subscription plans** — only pay-as-you-go, with real-time token-based billing.
+- Fixed default endpoint: `https://api.groq.com/openai/v1` (cannot be omitted).
+- For production, store API Key in `.security.yml` — avoid plaintext in `config.json`.
+- Inference is extremely fast; monitor usage to prevent unexpected overconsumption.