T09 · Insecure Skill Coding Practices
Warning
- Location
scripts/search.py:109- Finding
API Key Exposure Through Command-Line Input and Request Logging
- Content
View full analysis
- Remediation
View remediation
Security audit
baidu-maps-poi-ai-search
Security checks for vulnerabilities and agentic risk
Overview
The skill mostly does the advertised Baidu Maps search, but its script can expose the Baidu API key and private search terms in logs.
Review before installing. Use only a restricted Baidu Maps API key, prefer BAIDU_AK over passing keys in JSON, avoid sensitive location queries, and remove or redact the stderr request logging before using this in shared terminals, CI, or agent environments.
Vulnerability Patterns
- 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
- Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Findings (1)
Vulnerability Patterns
- Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
- MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
- MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
- Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
- Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (8)
Tp4
High
- Category
- MCP Tool Poisoning
- Confidence
- 91% confidence
- Finding
The code clearly matches part of the description: it searches for places/POIs via Baidu Maps and can include richer result fields when scope=2. However, the declared description explicitly says it can also retrieve detailed POI information by UID. No such UID lookup endpoint or function exists in the supplied code; UID only appears as a field in search results. Therefore the description overstates the implemented functionality. There are no obvious undeclared harmful capabilities, but there is a material mismatch because one of the core declared features is absent.
- Content
Undeclared Tool Scope
Medium
- Category
- MCP Least Privilege
- Confidence
- 92% confidence
- Finding
The skill requires environment access to
BAIDU_AKand performs network requests, but it does not declare an explicit tool scope such as permissions or allowed-tools. This weakens sandboxing and reviewability because consumers cannot clearly see that the skill will access secrets and transmit data externally, increasing the chance of unintended secret exposure or overly broad execution privileges.- Content
Natural-Language Policy Violations
Medium
- Category
- Not specified by scanner
- Confidence
- 91% confidence
- Finding
The natural-language documentation and usage guidance are presented in Chinese, but the file does not indicate that this locale is optional or that users can choose another language. Under the policy, forcing a specific language without user opt-in is a natural-language policy concern.
- Content
External Transmission
Medium
- Category
- Data Exfiltration
- Confidence
- 84% confidence
- Finding
The skill is designed to send user-supplied queries and use an API key to an external Baidu Maps endpoint. External transmission is expected for this skill's purpose, but it remains security-relevant because potentially sensitive location queries and secret-backed requests leave the local environment and go to a third party.
- Content
text ## API 文档 - 多维检索：`https://api.map.baidu.com/api_place_pro/v1/region`
Natural-Language Policy Violations
Medium
- Category
- Not specified by scanner
- Confidence
- 83% confidence
- Finding
The file presents itself entirely as a Baidu Maps search script in Chinese and its documented API usage is tied to Chinese-language behavior, while not offering any user-facing language/locale choice. Under the policy rule, forcing a specific language or locale without opt-in is a natural-language policy concern unless clearly justified as region-specific.
- Content
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
python #!/usr/bin/env python3 """ Baidu Maps Place Search - 多维检索 API Docs: https://api.map.baidu.com/place/v3/multidimensional """ import sys
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
python #!/usr/bin/env python3 """ Baidu Maps Place Search - 多维检索 API Docs: https://api.map.baidu.com/place/v3/multidimensional """ import sys
Missing User Warnings
Medium
- Category
- Not specified by scanner
- Confidence
- 98% confidence
- Finding
The script logs the full parsed request body to stderr, which can include sensitive user queries and the 'ak' API key. In agent, server, or CI environments, stderr is often collected centrally, so this can leak credentials and potentially sensitive location/search data to operators, logs, or downstream systems.
- Content