Prior Openclaw

Security checks across malware telemetry and agentic risk

Overview

This appears to be the claimed Prior knowledge-search integration, but it should be reviewed because it can send frequent work context to a third-party service and encourages pasting and storing an API key through chat.

Install only if you are comfortable with Prior receiving search queries and selected task/error context. Prefer setting PRIOR_API_KEY yourself through a local environment variable or CLI rather than pasting it into chat, and review each search or contribution for secrets, file paths, customer data, or proprietary details before allowing it to be sent.

SkillSpector

By NVIDIA

Vulnerability Patterns

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

Findings (2)

Vague Triggers

Medium

Confidence: 88% confidence
Finding: The skill instructs the agent to search Prior for a very broad set of common situations, including routine setup, architecture choices, unfamiliar tools, and any failed attempt. This can cause frequent transmission of user task context, errors, and implementation details to an external service by default, expanding data exposure and making the third-party dependency unusually pervasive in normal workflows.

Ssd 3

Medium

Confidence: 95% confidence
Finding: The skill explicitly encourages the user to paste their API key into the chat so the agent can configure it automatically. Collecting secrets through conversational input increases the chance of credential leakage through logs, transcripts, prompt history, downstream tools, or accidental reuse by the agent.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal