Back to skill

Security audit

助餐点Skill

Security checks for vulnerabilities and agentic risk

Overview

ClawScan marks this skill suspicious because scanners found a conditional risk that its API key and location queries could be sent to an untrusted destination, and the scan was partial.

Consider the scanner’s recommendation not to install until the endpoint is confirmed to use HTTPS and an approved host, and users are clearly told what location data is sent and to whom. The scan was partial: one Python file received only partial analysis because a parser limit was reached, so the evidence is incomplete.

Vulnerability Patterns
  • 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
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • Taint TrackingDirect Taint Flow, Variable-Mediated Taint Flow, Credential Exfiltration Chain
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (3)

Tainted flow: 'req' from os.environ.get (line 35, credential/environment) → urllib.request.urlopen (network output)

Critical
Category
Data Flow
Confidence
90% confidence
Finding

Sending a purpose-specific API key as authentication is legitimate, but the destination comes entirely from GAODE_POI_BASE without HTTPS or trusted-host validation. An insecure or attacker-influenced configuration can disclose the key and location queries, and default redirect handling can forward the custom X-API-Key header to another host; this is a conditional credential-disclosure risk, not evidence of deliberate exfiltration.

Content

Scanner excerpt · govt_meal_skill.py (reported line 40)May include surrounding context.

python
"User-Agent": "govt-meal-skill/1.0.0",
    })
    try:
        with urllib.request.urlopen(req, timeout=15) as r:
            return json.loads(r.read().decode())
    except urllib.error.HTTPError as e:
        body = e.read().decode(errors="replace")

Skill declares no tool scope ('permissions' or 'allowed-tools') but code capabilities were detected: env, network.

Medium
Category
MCP Least Privilege
Confidence
70% confidence
Finding

Without declared permissions the skill's intent is opaque and cannot be validated.

Content

No source excerpt is available for this finding.

The location-based query workflow lacks a privacy notice about external processing of the supplied reference location.

Medium
Category
Output Handling
Confidence
94% confidence
Finding

The workflow requests a reference location without explaining that it will be processed by an external service. A supplied home address or nearby landmark can reveal an elderly person's whereabouts, and the emphasis on government-sourced listings may obscure the distinction between public listing data and privately supplied query data.

Content

No source excerpt is available for this finding.