ClawHub
Back to skill

Security audit

boardraw

Security checks across malware telemetry and agentic risk

Overview

The skill is mostly aligned with generating editable diagrams, but it defaults toward uploading diagram content to boardraw.com and using an API key with under-scoped disclosure and credential handling.

Review this skill before installing if you only want local diagrams. It can create useful editable Excalidraw files, but its normal workflow uploads the generated board to boardraw.com, uses a BOARDRAW_API_KEY, and may look for that key in parent-directory .env files. Avoid using it for sensitive architecture, business, or personal diagrams unless you are comfortable sending that content to Boardraw and using the configured account key.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
Findings (9)

Lp3

Medium
Category
MCP Least Privilege
Confidence
94% confidence
Finding
The skill uses sensitive capabilities—environment access, local file read/write, and network transmission—without declaring them. That prevents proper consent and sandbox policy review, and in this skill those capabilities are used to read credentials and upload user-generated content externally.

Tp4

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding
The stated purpose is local diagram generation, but the instructions also include reading API keys and uploading whiteboard data to an external service. This is a material behavior mismatch that can mislead users and security controls about data egress and credential handling.

Description-Behavior Mismatch

Medium
Confidence
92% confidence
Finding
The documentation initially frames the skill as producing a local editable Excalidraw file, but later directs the agent to upload that content and return a hosted link. This inconsistency can cause unintended external sharing of user content, especially if users believe output remains local.

Context-Inappropriate Capability

Medium
Confidence
95% confidence
Finding
The skill instructs searching environment variables or walking up directories for a .env file to obtain API credentials, which expands credential access beyond what is minimally necessary. Broad .env discovery risks exposing unrelated secrets in shared workspaces or repositories.

Vague Triggers

Medium
Confidence
84% confidence
Finding
The invocation rules are very broad and can trigger on generic drawing or mapping requests, increasing the chance the skill runs in contexts where external upload, file creation, or credential access is unnecessary. Overbroad activation raises the attack surface and the likelihood of unintended data handling.

Missing User Warnings

High
Confidence
96% confidence
Finding
The workflow instructs uploading user-generated whiteboard content to boardraw.com without a prominent privacy or data-transfer warning at the decision point. Users may include sensitive architecture, org, or process information and not realize it will be sent to a third-party service.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The instructions to obtain credentials from env or .env files do not warn users that the skill will access sensitive secrets. In combination with network upload behavior, this creates a risk of silent secret usage without informed consent.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
This code sends the full whiteboard contents and the API key to boardraw.com, but the script provides no explicit privacy notice, confirmation prompt, or data-minimization guard before transmission. In a skill that may be used on user-provided diagrams containing sensitive business, architectural, or personal information, silent upload increases the risk of unintended external disclosure.

Credential Access

High
Category
Privilege Escalation
Content
1. 登录 boardraw.com → 设置 → API Keys
2. 创建新的 API Key（需要 Pro 或 Team 计划）
3. 添加到项目根目录的 .env 文件：
   BOARDRAW_API_KEY=br_your_api_key_here

   或者设置环境变量：
Confidence
89% confidence
Finding
The skill explicitly directs use of a .env file for API keys, confirming credential access behavior. While not inherently malicious, handling secrets inside a broadly-invoked skill increases the risk of accidental exposure, misuse, or over-collection if runtime boundaries are weak.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

Detected: suspicious.exposed_secret_literal

File appears to expose a hardcoded API secret or token.

Critical
Code
suspicious.exposed_secret_literal
Location
scripts/upload.py:33