A specialized utility to bypass OpenClaw's image delivery pipeline and send photos directly to Telegram via the Bot API.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install telegram-direct-send
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install telegram-direct-send using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is telegram-direct-send?
The telegram-direct-send skill is a high-reliability communication bridge designed specifically for developers using Openclaw Skills who need to ensure visual data reaches Telegram clients without fail. By bypassing the native Telegram gateway, which can occasionally suffer from silent image delivery failures (such as Issue #63137), this skill uses direct HTTP POST requests via curl to interact with the Telegram Bot API.
This approach provides a robust alternative for agents that need to report visual results, such as screenshots or data visualizations, directly to a user or group. It ensures that media is delivered instantly and correctly formatted, maintaining the integrity of your automated workflows within the Openclaw Skills ecosystem.
telegram-direct-send Use Cases
- Sending automated screenshots from an AI agent to a developer group.
- Bypassing known delivery issues in the standard OpenClaw Telegram channel.
- Delivering uncompressed high-quality documents and logs to mobile clients.
- Integrating interactive Telegram buttons into agent notification workflows.
How telegram-direct-send Works
- The skill identifies the necessary environment variables for the Telegram Bot Token and the Target Chat ID.
- It utilizes the curl binary to construct a multipart/form-data POST request.
- The request targets the official Telegram Bot API endpoints (e.g., sendPhoto or sendDocument).
- Files are uploaded directly from the local skill directory or workspace path to Telegram's servers.
- The Telegram Bot API processes the request and delivers the media to the specified chat or group instantly.
telegram-direct-send Setup
First, ensure you have curl installed. Then, configure your environment variables for your Openclaw Skills workspace:
# Copy the example environment file
cp .env.example .env
# Edit .env with your specific Telegram credentials
nano .env
# Source the variables into your current shell
source .env
To automate this for OpenClaw sessions, add the source command to your ~/.bashrc or create the send-image.sh wrapper script provided in the skill repository.
telegram-direct-send Data Schema & Taxonomy
The skill manages configuration through environment variables and local file paths.
| Variable | Type | Description |
|---|---|---|
| TELEGRAM_BOT_TOKEN | String | Your unique API token from @BotFather |
| TELEGRAM_CHAT_ID | String/Int | The ID of the recipient chat or group |
| .env | File | Local storage for secrets (excluded via .gitignore) |
| send-image.sh | Script | Optional wrapper for executing sends with local env loading |
telegram-direct-send Advanced Features
- Support for sendDocument to avoid Telegram's native image compression.
- Inline keyboard support for creating interactive bot responses.
- Integration with system date/time for dynamic captioning.
- Support for sendMediaGroup to handle multiple image uploads in a single transaction.
SKILL.md
Loading
Related Openclaw Skills
Screenshot → Telegram (Direct)
A high-performance utility to capture website screenshots and deliver them directly to Telegram chats via API, bypassing standard agent delivery bugs.
C-Suite Coach Video Marketing Skill
An AI-powered video marketing framework for coaching firms and L&D consultants to create high-authority content that converts enterprise decision-makers.
Proginn Developer Freelance Hub
A comprehensive tool designed to help developers manage profiles, bid on projects, and secure remote work opportunities on the Proginn platform.
User Feedback Claw
An automated AI agent skill that aggregates and analyzes user feedback across platforms to extract actionable product improvement signals.
Obsidain Karpathy Wiki
A professional-grade skill for managing an Obsidian vault as a persistent, compounding knowledge base using a structured raw-to-source-to-wiki pipeline.
Learn from Experience
A powerful self-learning skill that enables AI agents to evaluate their work, catch mistakes, and persist improvements across sessions through a tiered memory architecture.
Featured*
