Security audit

api-free

Security checks for vulnerabilities and agentic risk

Overview

This looks like an API reference skill, but it asks for broader execution and write authority than its documentation-only purpose supports.

Review before installing. This skill appears intended as a documentation helper, but it should not need write access, and webhook guidance is incomplete for secure production integrations. Use it only for high-level reference unless the publisher narrows the tool permissions and supplies the missing referenced API files or clear scope limits.

SkillSpector

Vulnerability Patterns

Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
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
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (4)

Description-Behavior Mismatch

High

Confidence: 94% confidence
Finding: The skill explicitly claims it is 'documentation-only' and will not execute user requests, yet the rest of the file describes use of exec-capable workflows, command-line reading via head/sed, and a live API-style output format. This mismatch can mislead operators and policy systems into granting trust to a skill that may actually perform actions, increasing the risk of unintended command execution or live handling of sensitive API inputs.

Intent-Code Divergence

Medium

Confidence: 80% confidence
Finding: The manifest says the skill should be used for webhook configuration, but the content later states webhook signature verification guidance is unavailable in the free version. That inconsistency can cause users or agents to configure webhooks without the required authenticity checks, which may lead to spoofed event acceptance or insecure integrations.

Context-Inappropriate Capability

Medium

Confidence: 86% confidence
Finding: Write capability is not justified by the stated purpose of a read-only API reference skill. Unnecessary write access enables tampering with local files, persistence of sensitive data, or modification of prompts/configuration if the skill is ever influenced by malicious input or chained with other tools.

Context-Inappropriate Capability

Medium

Confidence: 86% confidence
Finding: Write capability is not justified by the stated purpose of a read-only API reference skill. Unnecessary write access enables tampering with local files, persistence of sensitive data, or modification of prompts/configuration if the skill is ever influenced by malicious input or chained with other tools.

Static analysis

No suspicious patterns detected.