Security audit

Personal Client Management System & Finance System

Security checks across malware telemetry and agentic risk

Overview

This skill appears intended to manage an OurProject account, but it gives the agent broad API access to sensitive business and financial data and stores the API key insecurely.

Install only if you trust the publisher and intend to let your agent access OurProject finance, CRM, bills, tasks, and notifications. Use a least-privileged API key, keep the default HTTPS API URL unless you fully control the alternative, protect or remove .config.json when not needed, and review advanced API commands before allowing writes or account-changing requests.

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 (5)

Vague Triggers

Medium

Confidence: 88% confidence
Finding: The skill encourages broad natural-language requests like project, finance, CRM, bills, and notifications queries without defining guardrails around authorization, confirmation, or least-privilege use. In an agent setting, vague invocation patterns can cause overbroad API retrieval or unintended disclosure of sensitive workspace data when the user’s request is ambiguous.

Missing User Warnings

Medium

Confidence: 92% confidence
Finding: The documentation states the skill can access sensitive workspace, financial, CRM, bills, and notification data, but it does not warn users that API-backed retrieval may expose private business information through the agent interface. This increases the chance of accidental oversharing, unsafe deployment in shared environments, or use of overly privileged API keys without understanding the privacy implications.

Missing User Warnings

Medium

Confidence: 97% confidence
Finding: The script accepts a user-supplied API URL and deliberately falls back to the plain HTTP client when the URL is not HTTPS, then sends the bearer token in the Authorization header. That exposes the API key to interception or modification by any attacker on the network path, which can lead to full compromise of the user's OurProject account or workspace access tied to that key.

Missing User Warnings

Medium

Confidence: 89% confidence
Finding: The setup flow stores the API key in .config.json on disk in plaintext without warning the user or applying any file-permission hardening. If the host is shared, backed up, committed accidentally, or otherwise exposed, the credential can be recovered and reused to access the user's OurProject data and operations.

Missing User Warnings

Medium

Confidence: 91% confidence
Finding: The script pulls a broad cross-section of workspace data in one run, including projects, tasks, finance, bills, notifications, and CRM analytics, then prints a consolidated summary. In an agent skill context, this increases privacy exposure because a user may invoke a generic 'daily summary' without realizing the scope of sensitive data being accessed and surfaced.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal