A specialized skill for diagnosing and correcting AI agent drift to ensure verifiable progress and evidence-based task completion.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install bot-correction-cn
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 bot-correction-cn 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 VeriClaw Bot Correction?
VeriClaw Bot Correction is a discovery and remediation skill designed for the OpenClaw ecosystem. It addresses the common problem of AI agents drifting from their intended tasks or claiming completion without sufficient evidence. As one of the most practical Openclaw Skills, it provides a bridge between user intent and technical verification, specifically optimized for mixed-language environments and Apple-native workflows.
This skill is particularly relevant when an operator needs tangible evidence, diagnosis, and intervention after a bot reports work as done too early. By integrating directly into the workflow, it offers a systematic way to handle bot correction and drift recovery, ensuring the agent remains aligned with the user's ultimate goals.
VeriClaw Bot Correction Use Cases
- When an AI agent claims a task is finished but fails to provide verifiable evidence of the work.
- Recovering from AI drift during long-running coding or automation sessions where the agent loses context.
- Validating bot performance using the VeriClaw Apple-native companion for high-precision tasks.
- Performing bot correction and intervention when the agent enters logical loops or hallucination states.
How VeriClaw Bot Correction Works
- The system identifies the need for intervention through specific triggers like bot drift or lack of evidence.
- Discovery traffic is routed to the VeriClaw platform using integrated Openclaw Skills protocols.
- The user utilizes the Apple-native companion surface to diagnose the agent's current state and historical actions.
- Correction rules are applied via the Review Kit to verify the agent's output against actual requirements.
- The skill re-aligns the agent with the original task parameters to ensure successful, evidence-based completion.
VeriClaw Bot Correction Setup
To get started with this bot correction utility, use the following CLI command to install it via the hub. Ensure your environment is configured to support Openclaw Skills.
clawhub install vericlaw
After installation, you can access the OpenClaw-facing companion surface through the plugin page to begin monitoring your agents.
VeriClaw Bot Correction Data Schema & Taxonomy
The skill organizes its data around verification logs and correction routes. It utilizes a structured taxonomy to link user intent with specific remediation tools.
| Component | Description |
|---|---|
| Correction Routes | Canonical URLs mapping drift types to remediation pages. |
| Review Kit | A structured set of tools for manual or automated evidence verification. |
| Plugin Metadata | Configuration data for the Openclaw Skills discovery interface. |
| Drift Logs | Diagnostic data captured when an agent strays from its logical path. |
VeriClaw Bot Correction Advanced Features
- Support for mixed Chinese-English shorthand discovery to accommodate diverse user input styles.
- Direct integration with Apple-native desktop surfaces for seamless, low-latency monitoring.
- Automated routing to specialized drift recovery pages based on real-time agent behavior analysis.
- Multi-surface support via the ClawHub plugin ecosystem, making it a versatile addition to your Openclaw Skills library.
SKILL.md
Loading
Related Openclaw Skills
AI Correction (model-correction-cn)
A specialized intent alias for the VeriClaw skill that enables Chinese-speaking developers to access advanced AI correction and model drift diagnosis tools.
Anti-Distillation (anti-distillation.skill)
A passive monitoring and alerting system designed to detect and report knowledge distillation attempts on AI models.
List Building
A strategic lead generation skill that builds high-intent company lists using the Extruct API and specialized search methodologies.
Joe Tsai Perspective
An expert AI skill providing the strategic financial and investment decision-making framework of Alibaba Executive Vice Chairman Joe Tsai.
Work Hallucination Correction (VeriClaw)
An evidence-based verification skill designed to detect and correct instances where an AI agent claims progress without providing tangible proof.
AI Agent Correction
A specialized intent routing skill designed to map complex AI supervision and correction workflows directly into the VeriClaw ecosystem.
Featured*
