← Back to Skills Marketplace
eeoeofl

Digen Ai Free

by Eeofol · GitHub ↗ · v1.3.0 · MIT-0
cross-platform ⚠ suspicious
125
Downloads
1
Stars
0
Active Installs
15
Versions
Install in OpenClaw
/install digen-ai-free
Description
DigenAI image and video generation for OpenClaw. Supports image-to-video and text-to-image. Video generation via api.cowork.digen.ai with Bearer token. Trigg...
README (SKILL.md)

DigenAI Skill

Generate images from text prompts and videos from images via DigenAI API.

⚠️ First Time Users: Get Your Free API Key

Video generation requires a free API key (starts with ak_).

How to Get Your API Key

  1. Visit: https://claw.digen.ai
  2. Or join Discord: https://discord.gg/SRhbTt9hwp
  3. Or contact Telegram: @digen_skill_bot

Your API key is used as:

Authorization: Bearer YOUR_API_KEY

Note: The API uses https://api.cowork.digen.ai as the base URL.


Quick Start

from digen_ai_client import DigenAIClient

# Video generation — requires API Key (ak_xxx)
client = DigenAIClient(api_key="ak_xxxxxxxxxxxxxxxxxxxx")

Image Generation (Old API)

Uses api.digen.ai with DIGEN_TOKEN + DIGEN_SESSION_ID

Available Models

Model Description
default High quality model

Example

from digen_ai_client import DigenAIClient

client = DigenAIClient(
    old_api_token="your_token",
    old_api_session="your_session"
)

result = client.generate_image_sync(
    prompt="futuristic cyberpunk city at night, neon lights, rainy streets, highly detailed, 8K",
    model="default",
    resolution="1:1"
)

if result["success"]:
    print(f"✅ Image: {result['images'][0]}")
else:
    print(f"❌ Error: {result.get('error')}")

Video Generation (New API)

Uses new API with Bearer API Key

⚠️ Important: Use model="turbo"

The video generation API requires model="turbo" parameter (not default).

Available Models

Model Description Max Duration
turbo Fast and high quality generation 10s

Video Types

  • Image-to-Video: ✅ Works - requires image_url + prompt
  • Text-to-Video: ⚠️ May not work with all API keys (depends on credits)

Example: Image-to-Video (Recommended)

from digen_ai_client import DigenAIClient

client = DigenAIClient(api_key="ak_xxxxxxxxxxxxxxxxxxxx")

result = client.generate_video_sync(
    image_url="https://your-image.jpg",
    prompt="gentle camera pan left, neon lights twinkling",
    model="turbo",  # IMPORTANT: use "turbo", not "default"
    duration=5
)

if result["success"]:
    print(f"✅ Video: {result['video_url']}")
    print(f"   Thumbnail: {result['thumbnail_url']}")
else:
    print(f"❌ Error: {result.get('error')}")

Example: Text-to-Video (May Not Work)

client = DigenAIClient(api_key="ak_xxxxxxxxxxxxxxxxxxxx")

result = client.generate_video_sync(
    prompt="A cute cat playing piano in a cozy room, soft lighting",
    model="turbo",
    duration=5
)

if result["success"]:
    print(f"✅ Video: {result['video_url']}")
else:
    print(f"❌ Error: {result.get('error')}")
    # Note: Text-to-Video may fail if your API key only has image-to-video credits

API Key Management (New API)

Check API Key Info

client = DigenAIClient(api_key="ak_xxx")
info = client.get_api_key_info()
print(info)
# {'success': True, 'data': {'api_key': 'ak_xxx', 'status': 1, 'created_at': '...'}}

Upload Image

result = client.upload_image(file_path="/path/to/image.jpg")
if result["success"]:
    print(f"Image URL: {result['url']}")

Configuration

Environment Variables

Variable Description
DIGEN_TOKEN Old API token for image generation
DIGEN_SESSION_ID Old API session ID for image generation
DIGEN_API_KEY New API key (ak_xxx) for video generation

Setup

# Image generation (old API)
export DIGEN_TOKEN="your_token"
export DIGEN_SESSION_ID="your_session"

# Video generation (new API)
export DIGEN_API_KEY="ak_xxxxxxxxxxxxxxxxxxxx"

Error Handling

No API Key Error (Video)

❌ API Key Not Found!

Get your free API key:
- Visit: https://claw.digen.ai
- Or join Discord: https://discord.gg/SRhbTt9hwp

Error Codes (New API)

HTTP Code Error Code Description
400 invalid_request Invalid or missing parameters
401 invalid_api_key Invalid or missing API key
402 insufficient_credits Not enough credits
404 not_found Resource not found
500 internal_error Internal server error

API Reference

New API Endpoints (Video)

Method Endpoint Description
GET /b/v1/api-key Get API key info
POST /b/v1/upload Upload image file
POST /b/v1/video/generate Generate video
GET /b/v1/video/{id} Get video status

Base URL: https://api.cowork.digen.ai

Old API Endpoints (Image)

Method Endpoint Description
POST /v2/tools/text_to_image Generate image
POST /v6/video/get_task_v2 Get image status

Base URL: https://api.digen.ai


Scripts

  • scripts/digen_ai_client.py - Python client with sync/async support
  • scripts/batch_generate.py - Batch image generation utility
  • assets/telegram-bot.py - Telegram bot for API key distribution
  • assets/discord-bot.py - Discord bot for API key distribution

Tips

  • Video model: Always use model="turbo" (not default or seedance-2.0)
  • Image-to-Video: Requires image_url parameter
  • Text-to-Video: May not work with all API keys
  • Video generation: poll every 5 seconds, timeout 300s
  • Image generation: poll every 3 seconds, timeout 120s
Usage Guidance
This skill contains code that is consistent with a DigenAI client, but also includes Discord/Telegram bot scripts and hard-coded secrets. Before installing or running anything: - Do NOT run the provided run-*.sh scripts or launch the included bots without review. They contain hard-coded bot tokens and a MASTER_API_KEY which are sensitive and likely already compromised; assume these values are public and rotate any affected credentials. - Treat any embedded 'ak_*' keys or bot tokens in the package as leaked. If you control those accounts, revoke and rotate them immediately. Do not reuse those keys. - The skill metadata declares no required env vars, but the code expects DIGEN_API_KEY, DIGEN_TOKEN, DIGEN_SESSION_ID, plus bot tokens and a MASTER_API_KEY. If you only need the client, remove or ignore the bot assets and run only the digen_ai_client code; set API keys in your environment rather than using bundled defaults. - The bots persist user_keys.json locally; that file will contain generated API keys in plaintext. If you run a bot, store that file securely, restrict access, or use a secure backend for keys instead. - If you want to use the skill as an end user: ask the publisher to remove embedded credentials, update the skill metadata to declare required env vars, and avoid bundling live master tokens in distributed code. If the source is untrusted or you cannot get assurances, consider using only the client code after auditing it and supply your own API keys. Given the mismatches and embedded secrets, the package is suspicious and requires remediation or clarification from the publisher before safe use.
Capability Analysis
Type: OpenClaw Skill Name: digen-ai-free Version: 1.3.0 The skill bundle contains several high-risk security vulnerabilities and poor practices. Multiple shell scripts (assets/run-discord.sh, assets/run-telegram-bot.sh) contain hardcoded sensitive credentials, including a 'MASTER_API_KEY' and Discord/Telegram bot tokens. The core client logic in scripts/digen_ai_client.py explicitly disables SSL certificate verification (verify=False), exposing the agent to Man-In-The-Middle (MITM) attacks. Furthermore, the scripts contain absolute local file paths (e.g., /Users/mac/...), indicating the bundle was likely packaged from a live development environment without sanitization. While no clear evidence of intentional malice or data exfiltration was found, the inclusion of active credentials and insecure network configurations is highly irregular.
Capability Tags
cryptocan-make-purchasesrequires-oauth-tokenrequires-sensitive-credentials
Capability Assessment
Purpose & Capability
The SKILL.md describes an API client for DigenAI (image/video generation). Included Python client and scripts align with that purpose. However the package also contains Discord/Telegram bot code and shell run-scripts with embedded bot tokens and a MASTER_API_KEY. The registry metadata claims no required environment variables, but both the documentation and code expect/require DIGEN_API_KEY, DIGEN_TOKEN, DIGEN_SESSION_ID, DISCORD_BOT_TOKEN, TELEGRAM_BOT_TOKEN and a MASTER_API_KEY. Requiring a master API key and bot tokens is not strictly necessary for a simple client library or skill — the presence of key-distribution bots is additional functionality that should have been declared.
Instruction Scope
SKILL.md focuses on using the DigenAI client and setting DIGEN_API_KEY / DIGEN_TOKEN / DIGEN_SESSION_ID. It does not instruct the user to run the included bot scripts, but the repo contains run-discord.sh, run-telegram-bot.sh/run-tg.sh and bot implementations that will create and distribute API keys. The scripts set environment variables (including hard-coded tokens) and change directories to user-specific paths (/Users/mac/...), which is unexpected and brittle. The bots also persist user keys to a local user_keys.json file — collecting and storing generated keys locally, which increases attack surface and data exposure.
Install Mechanism
This is an instruction-only skill with no install spec. No downloads or extraction operations are present in the metadata. The risk comes from files included in the bundle (scripts and bots) rather than from an installer.
Credentials
The skill metadata declared no required env vars, but the code expects multiple secrets: DIGEN_API_KEY (new API), DIGEN_TOKEN and DIGEN_SESSION_ID (old API), DISCORD_BOT_TOKEN, TELEGRAM_BOT_TOKEN, and a MASTER_API_KEY used to mint per-user keys. Worse, a MASTER_API_KEY value and bot tokens are embedded directly in code and shell scripts. A master API key that can create other API keys is high-privilege; embedding it in the shipped code (and in run-*.sh) is disproportionate and dangerous. The skill also writes and reads a local user_keys.json containing per-user API keys — again sensitive.
Persistence & Privilege
The skill is not marked always:true and does not request platform-wide privileges. However, the included bot code is designed to run as persistent services (polling Discord/Telegram) and to persist data (user_keys.json) and may run outside the agent lifecycle if a user executes the run scripts. That persistent behavior is not described in the skill metadata and could have long-lived effects if someone runs the scripts.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install digen-ai-free
  3. After installation, invoke the skill by name or use /digen-ai-free
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.3.0
Added video_url_short field with TinyURL shortening for long S3 URLs.
v1.2.2
Updated Discord invite link to permanent URL.
v1.2.1
Image-to-video works with model=turbo. Updated landing page. Text-to-video note added.
v1.2.0
Video: use model=turbo. Image-to-video works. Text-to-video may fail with some keys.
v1.1.3
Removed api.cowork.digen.ai from client script (NEW_API_BASE now generic). Full sweep: no test.cowork.digen.ai or portal URLs remain anywhere in the bundle.
v1.1.2
Removed all test.cowork.digen.ai and api.cowork.digen.ai references from public SKILL.md and internal docs. API endpoints now genericized.
v1.1.1
Security cleanup: removed discord-send-messages.py (hardcoded token), telegram-bot.py (default token), .env. Deleted both bot scripts from bundle - client-only package now. SKILL.md description simplified.
v1.1.0
Removed model names and parameters from public-facing SKILL.md and landing page. Hidden: model names (flux2-klein/wan/turbo/2.6), aspect ratios, resolution details. External mystery preserved.
v1.0.6
Added Discord back alongside Telegram: Hero section now shows both TG Bot and Discord buttons; Community section has both Telegram and Discord links.
v1.0.5
Fixed install command in landing page: clawhub install digen-ai-free
v1.0.4
API key channel changed from Discord to Telegram bot @digen_skill_bot (send /key in DM or !key in group). Updated SKILL.md, client, and landing page.
v1.0.3
Fixed: unified API key instructions to Discord only (removed portal URL); cleaned up DIGENAI_TOKEN fallback; removed portal-link CSS; landing page now consistently points to Discord #api-key channel
v1.0.2
Full English SKILL.md; API key instructions updated to Discord #api-key channel (https://discord.gg/4shAQahd)
v1.0.1
Clean bundle: removed Discord bot files, updated to new API (video) + old API (image) architecture, added API key management endpoints
v1.0.0
- Initial release of DigenAI integration for OpenClaw, enabling image and video generation via DigenAI APIs. - Supports image generation through the old API (requires DIGEN_TOKEN and DIGEN_SESSION_ID). - Enables video generation via the new API (requires free Bearer API key obtained from https://test.cowork.digen.ai/b/portal). - Provides Python client scripts for both image and video generation, including sync/async support and batch tools. - Detailed documentation includes available models, usage examples, error handling, and environment variable setup.
Metadata
Slug digen-ai-free
Version 1.3.0
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 15
Frequently Asked Questions

What is Digen Ai Free?

DigenAI image and video generation for OpenClaw. Supports image-to-video and text-to-image. Video generation via api.cowork.digen.ai with Bearer token. Trigg... It is an AI Agent Skill for Claude Code / OpenClaw, with 125 downloads so far.

How do I install Digen Ai Free?

Run "/install digen-ai-free" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Digen Ai Free free?

Yes, Digen Ai Free is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Digen Ai Free support?

Digen Ai Free is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Digen Ai Free?

It is built and maintained by Eeofol (@eeoeofl); the current version is v1.3.0.

💬 Comments