ClawHub

Better Ralph

Security checks across malware telemetry and agentic risk

Overview

This skill appears to automate a coding workflow, including branches, tests, commits, and progress files, which fits its stated purpose but requires normal repository-level caution.

Install only if you want an agent to make code changes, run project scripts, update PRD/progress files, and create git commits. Before use, review prd.json/progress.txt for untrusted instructions or shell-like values, and ask the agent to show the intended branch, story, commands, and commit message before it runs them.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Confidence
91% confidence
Finding
The metadata description embeds several broad natural-language trigger phrases such as 'run better ralph' and 'next prd story' that could plausibly appear in ordinary discussion or planning. Because this skill performs powerful actions like editing files, executing commands, switching branches, and committing changes, ambiguous invocation increases the risk of accidental activation and unintended repository modification.

Vague Triggers

Medium
Confidence
89% confidence
Finding
The 'When to Use' section repeats vague activation examples without defining boundaries or disallowed contexts, so an agent could interpret conversational references like 'next prd story' or 'ralph loop' as authorization to execute the workflow. In this skill, accidental execution is more dangerous than usual because the workflow includes code changes, shell commands, git operations, and state updates to prd.json and progress.txt.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal