Back to skill

Security audit

UI Prototype Generator

Security checks for vulnerabilities and agentic risk

Overview

The skill is mostly a normal UI prototype generator, but its Figma path reads local access-token files too broadly and its packaging script could accidentally include credentials.

Review this before installing if you plan to use Figma. Use a dedicated, revocable Figma token with minimal scope, keep real auth-profiles.json files outside the skill package directory, and avoid running the packaging script near credential files. Treat generated output paths as untrusted unless you control the requested names. HTML-only use is much lower risk.

Vulnerability Patterns
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
Findings (4)

T05 · Unauthorized Access and Privilege Escalation

Warning
Location
scripts/figma_generator.py:24
Finding

Unnecessary Automatic Discovery and Loading of Figma Credentials

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Error
Location
scripts/package_skill.py:35
Finding

Credential Files Can Be Accidentally Included in Distributable Skill Archives

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/html_generator.py:37
Finding

Unrestricted HTML Output Path and Unescaped Title Derived from User Input

Content
View full analysis
{title} ''' ``` ### Technical Analysis The command-line `--output` value reaches `output_name` without validation. It is then used in two security-sensitive contexts: 1. It is concatenated directly into a filesystem path. 2. It is interpolated directly into an HTML document without HTML escaping. An output name containing absolute-path or traversal components can cause the generated `.html` file to be written outside the intended working directory. Existing files whose resulting names end in `.html` can be overwritten because the script opens the destination in write mode. The same value is embedded in the `` element. If an attacker can provide a crafted value containing markup that exits the title context—and can satisfy any filesystem path constraints caused by separator characters—the resulting generated document ...[truncated 1543 chars]
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/figma_generator.py:105
Finding

Figma Prototype Name Allows Output Path Traversal and File Overwrite

Content
View full analysis
Remediation
View remediation
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
  • Rogue AgentSelf-Modification, Session Persistence
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
Findings (28)

Tp4

High
Category
MCP Tool Poisoning
Confidence
91% confidence
Finding

This mismatch explicitly mentions reading a local credential store (auth-profiles.json) even though that sensitive local-access behavior is not clearly declared in the skill's purpose. Undisclosed access to stored tokens materially changes the security posture because users may invoke a prototype skill without realizing it can inspect local secrets and use them for API operations.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
91% confidence
Finding

This mismatch explicitly mentions reading a local credential store (auth-profiles.json) even though that sensitive local-access behavior is not clearly declared in the skill's purpose. Undisclosed access to stored tokens materially changes the security posture because users may invoke a prototype skill without realizing it can inspect local secrets and use them for API operations.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
92% confidence
Finding

This mismatch explicitly mentions reading a local credential store (auth-profiles.json) even though that sensitive local-access behavior is not clearly declared in the skill's purpose. Undisclosed access to stored tokens materially changes the security posture because users may invoke a prototype skill without realizing it can inspect local secrets and use them for API operations.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

This mismatch explicitly mentions reading a local credential store (auth-profiles.json) even though that sensitive local-access behavior is not clearly declared in the skill's purpose. Undisclosed access to stored tokens materially changes the security posture because users may invoke a prototype skill without realizing it can inspect local secrets and use them for API operations.

Content

No source excerpt is available for this finding.

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · README.md (reported line 67)May include surrounding context.

md
"note": "Get your token from https://www.figma.com/settings"
    }
  },
  "_comment": "Replace YOUR_FIGMA_TOKEN_HERE with your actual Figma Personal Access Token"
}

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · auth-profiles.template.json (reported line 11)May include surrounding context.

json
"note": "Get your token from https://www.figma.com/settings"
    }
  },
  "_comment": "Replace YOUR_FIGMA_TOKEN_HERE with your actual Figma Personal Access Token"
}

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · scripts/figma_generator.py (reported line 58)May include surrounding context.

python
def check_auth(self):
        """Check if Figma authentication is available"""
        if not self.access_token:
            print("❌ Figma access token not found")
            print("\nTo set up Figma authentication:")
            print("1. Go to https://www.figma.com/settings")
            print("2. Scroll to 'Personal Access Tokens'")

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · scripts/figma_generator.py (reported line 61)May include surrounding context.

python
print("❌ Figma access token not found")
            print("\nTo set up Figma authentication:")
            print("1. Go to https://www.figma.com/settings")
            print("2. Scroll to 'Personal Access Tokens'")
            print("3. Click 'Create new token'")
            print("4. Copy the token (starts with 'figd_')")
            print("5. Save it to auth-profiles.json:")

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The README instructs users to use a third-party Figma API and supply an access token, but it does not clearly warn that uploaded screenshots, UI descriptions, and generated design structure may be transmitted to an external service. This creates a real privacy and data-governance risk because users may unknowingly send proprietary or sensitive design artifacts off-platform.

Content

No source excerpt is available for this finding.

Session Persistence

Medium
Category
Rogue Agent
Confidence
60% confidence
Finding

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Content

Scanner excerpt · README.md (reported line 71)May include surrounding context.

  1. Create auth-profiles.json:
    bash
    # Location: ~/.openclaw/agents/main/agent/auth-profiles.json
    {
    

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The README states that the skill can send user-provided design content to the Figma API and return a file link, but it does not clearly warn users that screenshots, descriptions, or other potentially sensitive UI data will be transmitted to a third-party service. In a prototyping skill, users may unknowingly include internal product screens, customer data, or unreleased designs, so the lack of an explicit disclosure meaningfully increases privacy and data-handling risk.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
90% confidence
Finding

The skill describes capabilities that imply file reads, file writes, and outbound network access, but it does not declare any explicit tool scope or allowed-tools boundary. This weakens least-privilege controls and makes it easier for a host agent to grant broader access than necessary, especially because the workflow also references credential reads and API calls.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
81% confidence
Finding

Natural-language policy violations include forcing a specific language or locale without user opt-in. The entire skill description and examples are in Chinese, and there is no indication that users may choose another language or that the skill is intentionally limited to a Chinese-speaking context.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The workflow describes reading stored credentials and sending prototype data to an external API without an explicit privacy or security warning. This is dangerous because users may provide proprietary screenshots, design details, or tokens without understanding that local secrets will be accessed and data will be transmitted off-system.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
91% confidence
Finding

This finding points to the concrete POST request that sends JSON payloads to the Figma API. In a design-prototype skill this behavior is legitimate, but without clear permission gating and data-handling disclosure it can still expose confidential UI assets or metadata to a third party.

Content

Scanner excerpt · SKILL.md (reported line 161)May include surrounding context.

  1. 调用 Figma API:
    python
    headers = {"X-Figma-Token": auth['access_token']}
    response = requests.post(
        "https://api.figma.com/v1/files",
        json={"name": "Prototype", "nodes": nodes},
        headers=headers
    

External Transmission

Medium
Category
Data Exfiltration
Confidence
91% confidence
Finding

This finding points to the concrete POST request that sends JSON payloads to the Figma API. In a design-prototype skill this behavior is legitimate, but without clear permission gating and data-handling disclosure it can still expose confidential UI assets or metadata to a third party.

Content

Scanner excerpt · SKILL.md (reported line 161)May include surrounding context.

  1. 调用 Figma API:
    python
    headers = {"X-Figma-Token": auth['access_token']}
    response = requests.post(
        "https://api.figma.com/v1/files",
        json={"name": "Prototype", "nodes": nodes},
        headers=headers
    

External Transmission

Medium
Category
Data Exfiltration
Confidence
89% confidence
Finding

The hardcoded external endpoint confirms third-party data transmission. While appropriate for optional Figma export, it becomes more dangerous in this skill because the same document also references local credential access and does not clearly separate default local HTML generation from optional networked behavior with consent safeguards.

Content

Scanner excerpt · SKILL.md (reported line 162)May include surrounding context.

python
   headers = {"X-Figma-Token": auth['access_token']}
   response = requests.post(
       "https://api.figma.com/v1/files",
       json={"name": "Prototype", "nodes": nodes},
       headers=headers
   )

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The package description is written entirely in Chinese, which indicates a language-specific experience without offering the user any language or locale choice. Under the policy, forcing a specific language without opt-in is a natural-language policy violation unless the locale constraint is clearly documented and justified.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The skill manifest describes generating interactive prototypes from images or descriptions, defaulting to HTML output and only using Figma when explicitly requested. This plugin manifest instead defines a Figma-side 'UI Prototype Importer' with no indication of HTML generation, suggesting the file implements only a Figma import path rather than the broader/default behavior claimed for the skill.

Content

No source excerpt is available for this finding.

Scope Creep

Medium
Category
Not specified by scanner
Confidence
82% confidence
Finding

This plugin manifest declares no permissions and explicitly forbids network access via allowedDomains set to 'none'. The skill description claims support for generating Figma design files through an API, which would ordinarily require network access or additional permissions, so this component’s declared permissions do not cover that advertised capability.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The top-level description says the tool 'Generates Figma designs via API from HTML or reference images', and the manifest states Figma files can be generated through an API. In practice, create_prototype explicitly avoids API calls and instead serializes a local JSON file for later manual import, so the implemented behavior is materially narrower than the claimed capability.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · scripts/figma_generator.py (reported line 21)May include surrounding context.

python
def __init__(self, access_token=None):
        self.access_token = access_token or self._load_token()
        self.base_url = "https://api.figma.com/v1"
        
    def _load_token(self):
        """Load Figma token from auth-profiles.json"""

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The docstring states 'Create or update Figma prototype', which implies direct creation or modification of a Figma resource. The implementation immediately notes that no real API call is made and instead generates a JSON file that can be imported later, contradicting the documented behavior.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The generated document sets <html lang="zh-CN">, which forces a specific locale in all output regardless of user preference or project context. This is a natural-language/locale policy issue because the skill does not offer opt-in, configuration, or a documented reason for restricting output to Chinese.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
84% confidence
Finding

SQP-3 适用于所有文件类型。该 README 全文以中文给出操作说明和使用方式,但没有说明这是可选本地化版本、也没有提供语言选择或用户 opt-in,可能与要求不强制特定语言的组织政策冲突。

Content

No source excerpt is available for this finding.