Clawshell
Alert level: High
Human-in-the-loop security layer. Intercepts high-risk commands and requires push notification approval.
Locality:Hybrid
Data access:Sensitive
Actions:Execute
Installs 5Downloads 390Stars 3Updated 229h ago
Why this rating
Deterministic checks triggered by the tool capabilities and evidence.
- Locality: Hybrid
Uses local shell execution with external notification channels (Pushover/Telegram).
- Data access: Sensitive
Command strings and outputs can include credentials, files, and operational secrets.
- Action surface: Execute
Runs arbitrary shell commands when allowed or approved.
Best practices
Follow these steps to reduce risk when using this skill.
- Keep notification tokens in secure env storage and rotate them regularly.
- Maintain strict blocklists for destructive commands in addition to default rules.
- Review `logs/clawshell.jsonl` regularly to audit approvals and risky command patterns.
Evidence links
Public sources backing the indicator assignments.
Always be careful when navigating away from the website.
Max-risk rule
If any capability reaches a higher level, the entire indicator level bumps up to keep ratings deterministic and easy to scan.