A specialized security utility that identifies how AI agent skills silently mutate and accumulate risks across multiple inheritance generations.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install evolution-drift-detector
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 evolution-drift-detector 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 Evolution Drift Detector?
Evolution Drift Detector is a critical security tool designed to solve the problem of silent mutation in the AI agent ecosystem. While a skill might pass an initial audit in its first version, successive forks and inheritance layers can introduce dangerous capabilities—such as unauthorized network access or external schema fetching—without triggering a new review. This tool ensures that the original verified safe status of Openclaw Skills remains valid by analyzing the entire lineage of a skill.
It bridges the gap between static analysis and evolutionary changes, allowing developers and security teams to maintain trust in their agentic workflows. By tracing every fork and modification, it provides a transparent view of how a simple file-utility might have evolved into a complex, potentially high-risk organism. This visibility is essential for maintaining the integrity of Openclaw Skills in production environments.
Evolution Drift Detector Use Cases
- Tracing the lineage of a skill to ensure it hasn't gained unauthorized network permissions through successive forks.
- Comparing two versions of the same skill to quantify semantic and functional drift over time.
- Auditing marketplace inheritance chains to determine if a verified badge is still applicable to the current generation.
- Identifying safety-reducing mutations where previous input validations have been removed for performance reasons.
How Evolution Drift Detector Works
- The detector performs lineage reconstruction by tracing the skill's inheritance markers back to the original source version.
- It executes a per-generation diff to identify exactly what capabilities, permissions, or dependencies were added at each fork point.
- A capability drift score is calculated using an algorithm that weights high-risk additions like HTTP requests more heavily than cosmetic changes.
- Each mutation is classified into categories such as functional expansion or safety reduction to provide context for the changes.
- A final report is generated with a re-audit recommendation based on whether the cumulative drift exceeds safety thresholds.
Evolution Drift Detector Setup
Ensure you have the necessary system dependencies installed to run the detector on your Openclaw Skills:
# Install required binaries
sudo apt-get install curl python3
To analyze a specific skill lineage within your environment:
# Run the drift detector against a skill slug
python3 evolution_drift_detector.py --slug data-sanitizer
Evolution Drift Detector Data Schema & Taxonomy
The tool organizes its findings into a structured Evolution Drift Report, tracking data across the following dimensions:
| Field | Description |
|---|---|
| Lineage Tree | A hierarchical map of all generations and forks from the root skill. |
| Drift Score | A numerical value (0-100) representing the total risk accumulation. |
| Mutation Tags | Classifications like cosmetic, functional, capability-expanding, or safety-reducing. |
| Audit Scope | Comparison between the original permitted actions and current actual capabilities. |
Evolution Drift Detector Advanced Features
- Integration with blast-radius-estimator to see how many active agents are affected by a drifted skill.
- Support for trust-decay-monitor to track both content-based and time-based audit validity for all Openclaw Skills.
- Automated mutation classification using heuristic analysis of code changes and external call detection.
- Custom drift alert thresholds to trigger CI/CD failures if inherited skills exceed specific security risk scores.
SKILL.md
Loading
Related Openclaw Skills
Economic Incentive Misalignment Detector
A strategic diagnostic tool that identifies structural marketplace pressures favoring publication volume over security and safety quality.
Delta Disclosure Auditor
A security-focused auditor that identifies transparency gaps in skill updates by verifying machine-readable deltas and chain-of-custody records.
Clone Farm Detector
A specialized tool for detecting coordinated clone campaigns and reputation gaming in AI agent marketplaces.
Capability Scope Expansion Watcher
A security auditing tool that identifies AI skills gradually claiming broader permissions through subtle, incremental version updates.
GEP Immune Auditor
A sophisticated security agent that serves as an immune system for AI ecosystems by scanning assets for malicious patterns and intent.
Hollow Validation Checker
A specialized diagnostic tool designed to uncover hollow or fake validation tests that falsely signal AI agent skill reliability.
Featured*
