Security audit

Http Api Test Runner

Security checks across malware telemetry and agentic risk

Overview

This API testing skill is mostly coherent, but it needs review because it combines real authentication material with runnable shell/network tests and an unrestricted response-save path.

Install only if you are comfortable with a skill that generates and runs shell scripts making HTTP requests to endpoints you provide. Use test or least-privilege credentials, avoid pasting production cookies or tokens directly into commands, inspect generated scripts before running them, and keep any expect.save paths inside a dedicated temporary output directory.

SkillSpector

By NVIDIA

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 (5)

Lp3

Medium

Category: MCP Least Privilege
Confidence: 91% confidence
Finding: The skill explicitly instructs the agent to generate files and execute shell commands (`bash -n`, `bash`, and environment-injected runs), which are code-capable behaviors. Because the skill has no declared permissions despite requiring file read/write and shell execution, it creates a mismatch between documented trust boundaries and actual capabilities, increasing the risk of unintended command execution, file modification, or misuse of user-supplied endpoints and secrets.

Description-Behavior Mismatch

Medium

Confidence: 95% confidence
Finding: The script honors a user-controlled `expect.save` path and writes the full HTTP response body to that location, creating parent directories if needed. Because the skill accepts user-provided test artifacts, this enables arbitrary file write within the permissions of the runner, which exceeds normal API verification behavior and could overwrite files, drop data in sensitive locations, or persist secrets from responses onto disk.

Missing User Warnings

Medium

Confidence: 93% confidence
Finding: The README explicitly tells users to supply a full Cookie header when running the test script, but provides no warning that cookies often contain live session tokens, CSRF tokens, or other sensitive authentication material. In a skill designed to build and run reusable HTTP test artifacts, this increases the chance that users paste production session data into shell history, logs, shared files, or generated artifacts, leading to credential leakage or session hijacking.

Missing User Warnings

Medium

Confidence: 93% confidence
Finding: The cookbook explicitly tells users to run the script with a real Cookie header and auth token, but provides no warning about secure handling, redaction, shell history exposure, or least-privilege test credentials. In an HTTP API testing skill, this is particularly risky because users are likely to paste production-like session material directly into terminal commands, increasing the chance of credential leakage through shell history, logs, screenshots, or process inspection.

Missing User Warnings

Medium

Confidence: 95% confidence
Finding: The example prompt explicitly asks users to provide complete authentication material, including full cookies or bearer tokens, without any guidance to redact, minimize scope, or use test credentials. In a skill designed to build and run HTTP API tests, this creates a realistic path for users to paste production secrets into artifacts, logs, or chat context where they may be retained or mishandled.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

Detected: suspicious.exposed_secret_literal

File appears to expose a hardcoded API secret or token.

Critical

Code: suspicious.exposed_secret_literal
Location: examples/auth-login-required/README.md:20