tencent-map

ISSUES FOUND max severity CRITICALSkillSpector Do not install 58 registry clawhub.ai last scanned 2026-02-01 1 scan

4 findings (1 critical, 2 high, 1 medium), mainly data exfiltration, command injection, resource abuse. Most recent scan 2026-02-01.

This Tencent Maps API integration skill presents CRITICAL and HIGH severity security risks. The primary concerns are: (1) Command injection vulnerabilities due to unvalidated user input in URL construction, (2) Unrestricted external API calls that could be abused for data exfiltration, (3) API key exposure risks, and (4) Unsafe processing of external API responses. While the skill's stated purpose (map API integration) is legitimate, the implementation lacks essential security controls including input validation, output sanitization, resource limits, and user confirmation for external data transmission. The skill requires immediate security improvements before it can be safely used, particularly around input validation and command injection prevention.
Scanner's overall assessment (LLM analyzer)

Findings4

Critical1

High2

Medium1

Low0

Analyzersstatic behavioral llm trigger meta

Second scanner: NVIDIA SkillSpector

VerdictDo not install 58

Score58 / 100

Issues5

CategoriesSupply Chain MCP Least Privilege

Version2.12.0 (static)

Independent static run, no language model, over the same files as the Cisco scan above. How the two scanners compare across the whole corpus: Two scanners, one corpus.

What this skill does

腾讯地图Web服务API集成，用于地点搜索、路线规划、逆地理编码等功能

Package: 5 files, 14 KB
SKILL.md: 132 words, 12+ sections
Code: Shell (1)
Requires: bins: curl python3; env: TENCENT_MAP_KEY
Version: 1.0.0
Homepage: https://lbs.qq.com/dev/console/application/

SKILL.md covers: 腾讯地图API技能 · 设置 · 功能 · 1. 地点搜索 (search) · 基础搜索 · 带分页搜索 · 指定区域搜索 · 2. 逆地理编码 (reverse_geocode) · 3. 地理编码 (geocode) · 4. 路线规划 (route) · 5. 周边搜索 (around) · 参数

Findings

CRITICAL Hardcoded API Key Exposure Risk via Environment Variable Data Exfiltrationllm

The skill requires the TENCENT_MAP_KEY environment variable to function. While using environment variables is better than hardcoding, the skill makes multiple external API calls to 'apis.map.qq.com' with this key. If the key is compromised or if the skill is used in an untrusted environment, the API key could be exposed through network traffic or logs. The skill does not implement any key validation, rotation, or protection mechanisms.

scripts/tencent_map.sh

API_KEY="$TENCENT_MAP_KEY"

if [ -z "$API_KEY" ]; then
    echo '{"error": "TENCENT_MAP_KEY环境变量未设置"}'
    exit 1
fi

Remediation: 1. Implement API key validation before use. 2. Add rate limiting to prevent key abuse. 3. Consider using a secure credential management system instead of plain environment variables. 4. Add logging controls to prevent key exposure in logs. 5. Document key rotation procedures.

HIGH Command Injection via Unvalidated User Input in URL Construction Command Injectionllm

The script constructs URLs by directly interpolating user-provided parameters (keyword, region, addresses) into curl commands without proper sanitization or URL encoding. An attacker could inject shell metacharacters or malicious payloads through these parameters, potentially leading to command injection. For example, a keyword containing backticks, semicolons, or other shell special characters could execute arbitrary commands.

scripts/tencent_map.sh

KEYWORD="$1"
REGION="${2:-广州}"
PAGE_INDEX="${3:-1}"
PAGE_SIZE="${4:-20}"

URL="${BASE_URL}/place/v1/search?boundary=region(${REGION},0)&keyword=${KEYWORD}&page_index=${PAGE_INDEX}&page_size=${PAGE_SIZE}&key=${API_KEY}&output=json"

Remediation: 1. Implement strict input validation for all user-provided parameters. 2. Use proper URL encoding for all query parameters before constructing URLs. 3. Validate that numeric parameters (page_index, page_size, coordinates) are actually numbers. 4. Sanitize string inputs to remove or escape shell metacharacters. 5. Consider using a safer method than direct string interpolation, such as curl's --data-urlencode option.

HIGH Unrestricted External Network Calls to Third-Party API Data Exfiltrationllm

The skill makes unrestricted HTTP requests to 'apis.map.qq.com' (Tencent Maps API) with user-controlled parameters. While this is the stated purpose of the skill, there are no safeguards against data exfiltration through these API calls. User-provided search terms, addresses, and location data are transmitted to external servers without user confirmation or data minimization controls. An attacker could use this to exfiltrate sensitive location information or probe internal network details.

scripts/tencent_map.sh:14

BASE_URL="https://apis.map.qq.com/ws"

# Multiple curl calls throughout the script:
curl -s "$URL"

Remediation: 1. Implement user confirmation before making external API calls with sensitive data. 2. Add data minimization controls to limit what information is sent. 3. Implement request logging for audit purposes. 4. Add rate limiting to prevent abuse. 5. Consider implementing a whitelist of allowed API endpoints. 6. Validate that responses come from expected domains.

MEDIUM Missing Resource Limits and Error Handling for API Calls Resource Abusellm

The script lacks proper resource limits, timeouts, and error handling for external API calls. There are no timeouts on curl commands, no limits on response sizes, and minimal error handling. This could lead to resource exhaustion if the API is slow to respond, returns extremely large responses, or if an attacker triggers many concurrent requests.

scripts/tencent_map.sh:38

RESPONSE=$(curl -s "$URL")

# No timeout, no max response size, no retry limits

Remediation: 1. Add timeout parameters to all curl commands (e.g., --max-time 30). 2. Implement maximum response size limits (e.g., --max-filesize). 3. Add retry limits and exponential backoff. 4. Implement proper error handling for network failures. 5. Add resource monitoring to detect abuse patterns. 6. Consider implementing circuit breaker patterns for API failures.

What is in the package

File	Kind	Size
`README.md`	doc	1.9 KB
`SKILL.md`	doc	2.1 KB
`USAGE_EXAMPLES.md`	doc	3.8 KB
`package.json`	config	543 B
`scripts/tencent_map.sh`	script	5.3 KB

External hosts referenced

lbs.qq.com apis.map.qq.com

Hostnames that appear in the skill's files. Referencing a host is not a finding; compare against the findings above and the skill's stated purpose.

Environment variables referenced

API_KEY BASE_URL KEYWORD KEYWORDS REGION REGIONS TENCENT_MAP_KEY

Package contents recorded 2026-10-22. The scan above ran against the version current at scan time; the two can differ if the skill has since been updated.

What we checked

Static analyzer. Pattern rules over SKILL.md and bundled files: known-bad commands, exfiltration URLs, secrets, obfuscation.
Behavioral analyzer. Checks what the skill asks the agent to do against what its description claims it does.
Llm analyzer. A language model reads the skill and reports intent-level issues the rules cannot express.
Trigger analyzer. Looks for instructions that fire on conditions the user did not ask for.
Meta analyzer. A second model pass that validates, deduplicates and re-scores the other analyzers' findings.

What we did not check

Runtime behaviour. The skill was not executed; findings come from reading its files.
Anything fetched at install or run time, including remote scripts, package dependencies, and URLs the skill tells the agent to open.
Content the analyzers cannot read: images, binaries, and some non-English text.
Known scanner bypass classes. See the methodology and limitations page.

Sources

Repository: https://clawhub.ai/skills/qqmap
Raw scan result: 腾讯地图api调用技能-scan.json
Badge for your README: ![skillscan.io](https://skillscan.io/badge/clawhub-ai-tencent-map.svg)
Request a re-scan