Security audit

Vibe Prospecting

Security checks across malware telemetry and agentic risk

Overview

This is a coherent B2B prospecting skill bundle with disclosed credential use, CLI installation, and CSV export behavior, but users should treat its data and API keys carefully.

Install this only if you intend to use Vibe Prospecting/Explorium for B2B data workflows. Protect the vpai config file and API keys, prefer controlled secret handling on shared or CI hosts, and use sample-first mode unless you are ready for a full export with the requested scale and privacy obligations.

SkillSpector

By NVIDIA

Vulnerability Patterns

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

Findings (5)

Missing User Warnings

Low

Confidence: 88% confidence
Finding: The document discloses that credentials are stored at a predictable path on disk without any warning about local secret storage risks. On shared machines, developer workstations, or CI runners, this can encourage insecure handling of cached credentials and increase the chance of accidental disclosure through backups, artifact collection, or lax file permissions.

Missing User Warnings

Medium

Confidence: 90% confidence
Finding: The skill instructs users to set an API key directly via `vpai config --api-key` and separately states that credentials are stored in `~/.config/vpai/config.json`, but it does not provide any warning about local secret storage, file permissions, shell history exposure, or safer secret-handling practices. In an agent workflow context, this can lead to long-lived credentials being written to disk or exposed in command history/logs, increasing the chance of credential theft and unauthorized API use.

Missing User Warnings

Medium

Confidence: 94% confidence
Finding: The skill instructs users to provide a tenant API key directly via command line or environment variable, but gives no warning about secure handling. Command-line arguments can leak via shell history, process listings, CI logs, and transcripts, making credential exposure plausible in exactly the terminal/automation contexts this file targets.

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: The skill directs the agent to copy the generated csv_path into the working directory, which creates or modifies local files as a side effect, but it does not require a user-facing warning or confirmation. Silent file creation can overwrite expected workspace contents, leak sensitive exported data into shared directories, or surprise users who did not consent to filesystem changes.

Missing User Warnings

Medium

Confidence: 98% confidence
Finding: This skill explicitly instructs the agent to bypass the normal 5-entity sample preview and approval gate, then execute the full prospecting workflow immediately. Removing an established confirmation checkpoint increases the risk of unintended large-scale actions, excessive data retrieval, cost overruns, or policy-violating enrichment without giving the user a clear warning at execution time.

VirusTotal

64/64 vendors flagged this plugin as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.