Oathe Security Badge

oathe / scan / reports / safe

Is 505768069-rgb/adaptive-skill-factory safe?

Yes. 505768069-rgb/adaptive-skill-factory is safe to install. Oathe's behavioral security audit gave the adaptive-skill-factory skill by 505768069-rgb a trust score of 86/100 with 3 findings, none critical or high. Report updated

https://clawhub.ai/505768069-rgb/adaptive-skill-factory

86
SAFE

Is 505768069-rgb/adaptive-skill-factory safe to install?

This skill provides legitimate functionality for managing and installing other skills from ClawHub. While it accessed sensitive credential files (triggering security honeypots), no actual data exfiltration occurred. The skill's ability to install other skills presents some risk but is used within appropriate guardrails.

What security issues were found in 505768069-rgb/adaptive-skill-factory?

Category Scores

Prompt Injection 85/100 · 30%
Data Exfiltration 75/100 · 25%
Code Execution 95/100 · 20%
Clone Behavior 90/100 · 10%
Canary Integrity 100/100 · 10%
Behavioral Reasoning 80/100 · 5%

Findings (3)

MEDIUM Sensitive File Access Detected -25 ▶

The skill accessed multiple sensitive credential files including SSH keys, AWS credentials, Docker config, and other authentication files during execution. While no actual data exfiltration occurred, this behavior indicates the skill or its dependencies are scanning the filesystem for sensitive data.

Evidence: Filesystem monitoring shows access to /home/oc-exec/.ssh/id_rsa, /home/oc-exec/.aws/credentials, /home/oc-exec/.env, /home/oc-exec/.npmrc, /home/oc-exec/.docker/config.json, /home/oc-exec/.config/gcloud/application_default_credentials.json

LOW Potential for Malicious Skill Installation -20 ▶

The skill's primary function is to help agents install other skills from ClawHub. While the skill itself appears benign and encourages installing 'the smallest relevant skill', it could potentially be manipulated to install malicious skills.

Evidence: SKILL.md contains instructions: 'If internet and ClawHub are available, search before building. Install the smallest relevant skill, not a broad bundle.'

INFO Legitimate ClawHub API Communications 0 ▶

The skill makes appropriate API calls to ClawHub servers for skill discovery and management functions. Network traffic patterns are consistent with legitimate skill management operations.

Evidence: Network connections to 216.150.1.1:443 with curl commands to 'https://clawhub.ai/api/v1/skills/adaptive-skill-factory'

Should I install 505768069-rgb/adaptive-skill-factory?

Oathe's verdict for 505768069-rgb/adaptive-skill-factory is SAFE with a trust score of 86/100. Recommendation: Install with caution.