A specialized skill designed to detect, tag, and log AI capability gaps and rule conflicts to drive continuous architectural evolution.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install capability-radar
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 capability-radar 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 Capability Radar?
Capability Radar acts as the essential feedback loop for AI agents, functioning as the first gear in an automated evolution engine. Instead of allowing errors or limitations to be buried in chat history, this skill captures tool failures, boundary reaches, and repetitive patterns in a structured format. It ensures that every instance where an agent struggles or relies on a manual workaround becomes a clear signal for improvement.
By systematically documenting these events within the Openclaw Skills ecosystem, the system can prioritize which new skills need to be developed or which internal rules require adjustment. This transforms the agent from a static tool into a self-improving entity that learns from its technical constraints and user interactions.
Capability Radar Use Cases
- Detecting when API tool calls fail or return error codes like 400.
- Identifying when an agent reaches its functional boundaries and cannot fulfill a request.
- Tracking repetitive requests that occur three or more times to signal a need for automation.
- Resolving rule conflicts where different instructions lead to contradictory agent behavior.
- Documenting manual workarounds that indicate a missing specialized skill.
- Capturing user corrections regarding context or identity to prevent future mistakes.
How Capability Radar Works
- The AI agent encounters a trigger event such as a tool failure, repeated request, or user correction.
- The agent invokes Capability Radar to structure the incident details.
- A standardized entry is appended to the daily memory file located at memory/YYYY-MM-DD.md.
- During session initialization, the Cowork engine scans these memory files for the specific radar tag.
- The engine evaluates the priority of logged items based on their impact and frequency.
- High-priority items trigger the generation of new skill drafts or rule modifications to close the capability gap.
Capability Radar Setup
To integrate this skill into your Openclaw Skills environment, ensure your agent has write access to the memory directory. No complex installation is required beyond adding the capability-radar definition to your skill bank. You can verify detected gaps using the following command:
grep "🛠 from-xiaomeng: Capability Radar" memory/*.md
Capability Radar Data Schema & Taxonomy
Capability Radar organizes its findings into a highly structured Markdown format within the memory logs. The schema includes the following fields:
| Field | Description |
|---|---|
| Time | The exact timestamp of the occurrence (YYYY-MM-DD HH:MM). |
| Trigger Type | Classification such as Tool Failure, Repeat Pattern, or Rule Conflict. |
| Scenario | A brief description of the context and what occurred. |
| Failure Point | The specific tool name or rule that caused the bottleneck. |
| Impact | Priority level (High, Medium, or Low) based on user experience. |
| Expected Capability | Description of the ideal behavior or missing functionality. |
| Suggested Action | Recommended next steps like creating a new skill or modifying AGENTS.md. |
Capability Radar Advanced Features
- Priority-based Evolution Protocol: Automatically categorizes gaps into Red (immediate fix), Yellow (weekly fix), or Green (backlog) tiers.
- Evolution Closed-Loop: Tracks the lifecycle of a gap from detection to the deployment of a new skill in the SkillBank.
- Pattern Recognition: Specifically monitors for the '3x Repeat Rule' to identify high-value automation targets.
- Conflict Resolution: Identifies contradictions between global agent frameworks and specific group rules.
SKILL.md
Loading
Related Openclaw Skills
Commitment Engine
A robust state-machine framework that converts conversational promises into persistent, trackable, and auto-executing files.
XiaoHongShu Content Migration
An automated solution for scraping XiaoHongShu notes and cross-posting them to Facebook Meta Business Suite and WordPress via REST API.
QMD Plus
QMD Plus enhances local markdown search by using LLMs to expand simple queries into sophisticated lexical and semantic search variations.
Harness Evolve
A self-evolution engine that consumes research logs and performs system self-checks to optimize AI agent architectures automatically.
Agent Harness Architect
A professional diagnostic skill designed to harden long-running AI agent architectures by implementing structured session bridges and verification protocols.
django-vue-admin Code Generator
An automated code generation tool that creates full-stack Django and Vue.js modules based on simple user descriptions.
Featured*
