Security audit

Agent News API Free Tier

Security checks for vulnerabilities and agentic risk

Overview

This skill is a small news-fetching CLI that only calls the disclosed Agent News API endpoint and does not read credentials, write files, or persist anything.

Installers should be comfortable with the skill sending query terms and limit values to agentnewsapi.com and receiving third-party news data. The scanner note is a documentation clarity issue, not evidence of hidden access or malicious behavior.

A.I.G

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

SkillSpector

Vulnerability Patterns

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
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (1)

Intent-Code Divergence

Low

Confidence: 79% confidence
Finding: The header says the version removes 'all environment variables' for compatibility and security, which implies avoidance of runtime-provided external inputs. However, the code immediately consumes process.argv to determine behavior and parameters. While argv is not the same as environment variables, the wording is actively misleading about what external process-level inputs are still used.

Static analysis

No suspicious patterns detected.