ClawHub

Self Improving Agent 1.0.0

Security checks across malware telemetry and agentic risk

Overview

This skill is not malicious, but it asks agents to automatically persist conversation-derived learnings and promote them into durable agent memory without enough privacy or approval guardrails.

Install only if you are comfortable with an agent writing persistent learning logs in the project. Before using it in a shared, proprietary, or version-controlled repo, require confirmation before logging conversation-derived content, redact secrets and personal or confidential data, and manually review any proposed changes to `CLAUDE.md` or `AGENTS.md`.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill directs automatic logging of user corrections, feature requests, and related interaction details into project files without any consent, minimization, or privacy screening step. This can persist sensitive user-provided content, internal prompts, filenames, environment details, or error output into the repository or workspace, creating an avoidable data retention and disclosure risk.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The promotion workflow encourages copying learnings into long-lived memory files like `CLAUDE.md` and `AGENTS.md` but does not warn against persisting sensitive or user-derived information there. Because these files are meant to be durable and broadly consulted, accidental promotion can amplify exposure of secrets, private data, or confidential project details beyond the original context.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal