A sophisticated conversational enhancement skill that transforms robotic AI responses into natural, human-like dialogue through intelligent tone and greeting management.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install human-like-reply
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 human-like-reply 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 Human-like Reply?
Human-like Reply is a specialized enhancement for AI agents designed to bridge the gap between machine-generated text and authentic human conversation. By analyzing dialogue history, topic shifts, and user behavior, this skill dynamically adjusts the AI's tone, phrasing, and structure. It effectively eliminates the repetitive, overly formal, and mechanical patterns often found in standard LLM outputs.
As part of the Openclaw Skills ecosystem, this tool provides a layer of emotional intelligence to interactions. It manages the frequency of greetings, introduces natural filler words, and varies sentence structures to ensure the AI sounds more like a real colleague or assistant rather than a static interface. Whether for casual chatting or professional support, it creates a more fluid and comfortable user experience.
Human-like Reply Use Cases
- Reducing robotic repetition in long-form chat sessions by scaling back formal greetings over time.
- Mirroring user behavior, such as responding with a natural greeting only when the user initiates one.
- Softening technical or dry information with colloquial expressions and appropriate emojis.
- Automatically resetting conversation tone and context after periods of silence or major topic changes.
- Adjusting the formality level of the agent dynamically based on specific project requirements.
How Human-like Reply Works
- The skill intercepts the raw AI response before it is sent to the user.
- It checks the conversation state to determine if a greeting is necessary based on the greeting_threshold and silence_minutes.
- A pattern-matching engine identifies and removes mechanical opening phrases like "Certainly," or "Okay,".
- Casual replacements are applied to formal terms (e.g., changing "Understood" to "Got it").
- Randomly inserts filler words and emojis based on pre-configured probability settings to simulate human hesitation and emotion.
- Adjusts punctuation and applies reduplication (e.g., "take a look" to "look-look") for a softer, more natural cadence.
- The final formatted response is delivered to the user interface.
Human-like Reply Setup
To install and test the Human-like Reply skill within your environment, use the following commands:
# Navigate to the skill directory
cd ~/.openclaw/workspace/skills/human-like-reply
# Run the diagnostic formatter test
python3 scripts/test_formatter.py
# Monitor the real-time conversation state
cat memory/reply_state.json
Human-like Reply Data Schema & Taxonomy
The skill maintains a persistent JSON state to track conversation dynamics and ensure continuity across sessions.
| Attribute | Description | Type |
|---|---|---|
rounds |
Number of exchanges in the current session | Integer |
last_message_time |
Timestamp of the last user interaction | ISO8601 String |
current_topic |
Keywords identifying the active subject | String |
greeting_count |
Number of times a formal greeting has been used | Integer |
last_used_greeting |
Tracks if the last response included a greeting | Boolean |
Human-like Reply Advanced Features
- Multi-level formality control (formal_level) to switch between casual and professional personas.
- Topic change detection that triggers a reset of conversational intimacy when the subject shifts.
- Time-aware greeting patterns that recognize morning, afternoon, and specific honorifics.
- Customizable filler_probability and smiley_probability for granular control over agent personality.
- Mirroring logic that adapts the AI's greeting style to match the user's input style.
SKILL.md
Loading
Related Openclaw Skills
Local Deep Research
An iterative research skill that interfaces with a locally-hosted LDR service to provide deep, cited reports without sending data to third-party clouds.
Accessory Bundle & Cross-Sell Designer
A strategic merchandising skill designed to help multi-SKU accessory brands increase AOV through automated bundle structures and high-conversion cross-sell logic.
QClaw
A specialized signal extraction skill designed to transform noisy WeChat group chat logs into actionable intelligence and executive summaries.
Stock Insight
A high-signal AI analysis skill that transforms raw market data into actionable professional investment reports using a structured three-stage research framework.
Ariadne Thread
Ariadne Thread creates AI-friendly project structures using a 4-level progressive disclosure indexing system to minimize token usage and maximize modification safety.
Changelog Generator
A powerful automation skill that transforms Conventional Commits into professional, auditable release notes and semantic versioning logic.
Featured*
