Security audit

amap-jsapi-skill

Security checks for vulnerabilities and agentic risk

Overview

The skill is mostly an AMap JSAPI helper, but it asks for broad read/write/exec authority and advertises unrelated security scanning and generic command/file capabilities without clear limits.

Review this skill before installing. It may be useful for AMap JSAPI examples, but grant read/write/exec access only if you are comfortable with a broadly scoped skill; keep AMap keys out of source code, prefer backend proxying for securityJsCode, and do not rely on this artifact for vulnerability scanning or compliance work.

SkillSpector

Vulnerability Patterns

Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration

Findings (5)

Description-Behavior Mismatch

Medium

Confidence: 94% confidence
Finding: The skill is presented as an AMap JSAPI development helper, but the documentation advertises unrelated capabilities such as deep vulnerability scanning, CVE correlation, compliance auditing, and asset risk scoring. This scope expansion can cause an agent to invoke the skill for security-sensitive tasks it is not designed or constrained to perform, increasing the chance of over-privileged use and unsafe handling of unrelated data.

Description-Behavior Mismatch

Medium

Confidence: 92% confidence
Finding: The documentation broadens the skill from map API usage into generic file processing, command execution, and information retrieval. For an agentic system, this creates dangerous ambiguity: a caller may treat the skill as a general-purpose execution utility, enabling access patterns and behaviors far beyond the declared map-integration purpose.

Context-Inappropriate Capability

Medium

Confidence: 95% confidence
Finding: Advertising system command execution in a map JSAPI development skill is unjustified and materially increases risk because command execution is a powerful primitive that can be repurposed for local discovery, secret access, or unintended side effects. In the context of an agent platform that exposes an exec tool, this mismatch makes prompt abuse and privilege misuse more dangerous.

Context-Inappropriate Capability

Medium

Confidence: 93% confidence
Finding: Claiming unrelated security scanning and asset risk scoring capabilities within a map-development skill is an unjustified expansion into sensitive security operations. This can mislead orchestration logic or users into routing privileged security analysis tasks through a skill that lacks clear scope boundaries and may expose sensitive assets or results unnecessarily.

Vague Triggers

Medium

Confidence: 88% confidence
Finding: The invocation language is overly broad, stating the skill should be used for API integration, interface docking, webhook configuration, and system connection scenarios generally. Such vague matching criteria can cause the skill to be selected for many unrelated tasks, especially in agent routing systems, increasing the chance that privileged tools are used outside intended scope.

Static analysis

No suspicious patterns detected.