A visual communication framework for AI agents to deliver scannable, high-signal messages across popular chat platforms.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install agent-format
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 agent-format 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 Agent Format?
Agent Format is a specialized set of patterns designed to standardize how AI agents communicate with human users. By moving away from dense walls of text and towards a structured visual hierarchy, this addition to your Openclaw Skills library ensures that critical status updates, progress indicators, and data trends are immediately recognizable. It leverages specific Unicode characters and emojis to create rich, platform-agnostic UI elements within chat interfaces like Slack, Discord, and WhatsApp.
The primary purpose is to improve human-agent interaction by assuming the user will scan rather than read every word. By utilizing standard formatting for progress bars, sparklines, and status indicators, developers can ensure their Openclaw Skills provide high-value, actionable information at a glance.
Agent Format Use Cases
- Monitoring long-running background tasks with visual progress bars to see completion percentages.
- Sending daily project status reports that use color-coded emojis to highlight blockers and urgent tasks.
- Displaying time-series data or performance trends using compact, single-line sparklines within a chat message.
- Organizing multi-column data into scannable tables using monospace formatting for consistency across mobile and desktop devices.
How Agent Format Works
- The AI agent evaluates the information to be sent and selects the appropriate visual pattern (e.g., a progress bar for tasks or a sparkline for trends).
- Status indicators (🟢, 🟡, 🔴) are prepended to headers to provide immediate context through color-coded signaling.
- The skill uses standardized Unicode blocks to construct layout elements like dividers and progress bars that render correctly across different operating systems.
- Messages are structured following a visual hierarchy: lead with the most important signal, followed by visual bars, and then detailed text to support the Openclaw Skills ecosystem.
Agent Format Setup
To integrate these patterns into your AI agent's output, include the Agent Format reference in your agent's system prompt or output instructions.
# No heavy installation required; these are formatting standards.
# Add the reference patterns to your Openclaw Skills prompt library.
Ensure your target chat platform supports Markdown and Monospace fonts for the best visual results.
Agent Format Data Schema & Taxonomy
The skill utilizes a set of standardized characters and symbols to organize information. This ensures that every message sent via Openclaw Skills maintains a consistent taxonomy.
| Element | Visual Chars | Purpose |
|---|---|---|
| Progress Bars | █ (filled), ░ (empty) |
Visual tracking of completion percentage |
| Sparklines | ▂ ▃ ▄ ▅ ▆ ▇ █ |
Representing data trends in a single line |
| Status Symbols | 🟢, 🟡, 🔴, 🔥, ⚡ | Signaling operational states (Done, WIP, Error, Urgent) |
| Monospace Tables | ─ (dividers) |
Aligning data columns for easy scanning |
Agent Format Advanced Features
- Cross-platform compatibility ensuring consistent visuals on Telegram, Discord, Slack, and WhatsApp.
- Dynamic progress bar generation that can be updated as the agent completes sub-tasks.
- Signal-first prioritization, automatically placing critical alerts at the very top of the message stream.
- Monospace-aligned table rendering to present complex data without requiring external spreadsheet links in your Openclaw Skills workflow.
SKILL.md
Loading
Related Openclaw Skills
Zotero
A comprehensive integration for managing Zotero personal and group libraries through AI-driven automation.
Hour Meter
A digital analog to the Hobbs meter for tracking elapsed time, milestones, and life events with tamper-evident verification.
Fastest Browser Use
A high-performance Rust-based browser automation engine designed for lightning-fast DOM extraction and efficient AI agent interactions.
Duo: AI Relationship Matchmaking
A relationship-focused matchmaking client for the NDAI Zone that leverages local profile data for secure, autonomous compatibility negotiations.
Board of Advisors — Strategic Counsel AI
A specialized AI framework featuring 15 virtual advisors providing high-level strategic counsel for company-defining decisions.
Board of Mentors: Domain Expert AI Advisors
A collection of 17 specialized AI mentor personas designed to provide strategic wisdom and domain-specific guidance for founders and operators.
Featured*
