Back to skill

Security audit

neon-ai-gateway

Security checks for vulnerabilities and agentic risk

Overview

ClawScan rates this skill suspicious with medium confidence, mainly because it found a possible privacy and access-scope issue in an example that sends database records to hosted inference; the scan was partial.

The verdict is suspicious, and Skillspector recommends against installing. Before deciding, consider whether you trust the Neon CLI command that manages skills and whether the example’s database access is restricted to the right user and necessary fields. The scan was partial: a static analysis step was degraded, no files were fully inspected, and several references could not be resolved, so the evidence is incomplete.

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
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • YARA SignaturesMalware Match, Webshell Match, Cryptominer Match
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (5)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
90% confidence
Finding

The supplied chunk contains only a skills-management command. It does not call models, generate embeddings, configure an AI Gateway client, or route AI requests. Although tooling setup can support the declared purpose, this chunk shows no connection between the skills operation and that purpose. The precise effects depend on the installed Neon CLI.

Content

No source excerpt is available for this finding.

YARA rule 'agent_skill_mcp_tool_poisoning_metadata': MCP/tool metadata poisoning indicators in tool schemas or skill manifests [agent_skills]

High
Category
YARA Match
Confidence
80% confidence
Finding

YARA rule matched a hack tool or exploit indicator (offensive tools, reconnaissance, privilege escalation, or exploit frameworks).

Content

Scanner excerpt · SKILL.md (reported line 3)May include surrounding context.

md
---
name: neon-ai-gateway
description: >-
  One API and one credential for frontier and open-source LLMs and
  embedding models, built into your Neon branch and powered by Databricks.
  Use when a user wants to call an LLM, generate embeddings, add AI/chat/an
  agent to their app, route between model providers (OpenAI, Anthropic,
  Google/Gemini, Meta, Alibaba, and more), or avoid juggling separate
  provider API keys and accounts — especially when they already use Neon
  and want AI requests to branch with their project. Works with the OpenAI
  SDK, Anthropic SDK, google-genai, the Vercel AI SDK, and Mastra by
  cha

Hidden Instructions

High
Category
Prompt Injection
Confidence
70% confidence
Finding

Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 326)May include surrounding context.

Response shape — OpenAI/OpenRouter-compatible list:

jsonc
{
  "object": "list",

Skill Enumeration

Medium
Category
Agent Snooping
Confidence
80% confidence
Finding

Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.

Content

Scanner excerpt · SKILL.md (reported line 25)May include surrounding context.

FIRST: Use the parent neon skill for a Neon overview, getting started with Neon, Neon development best practices, and more.

If the neon skill is not installed, fetch it from https://neon.com/docs/ai/skills/neon/SKILL.md or install it with:

bash
neon skills -s neon -y

The database-backed agent example sends user todo data to hosted inference without a privacy warning.

Medium
Category
Output Handling
Confidence
90% confidence
Finding

Hosted inference is the skill's explicit purpose, so the absence of a privacy warning alone does not establish unauthorized disclosure. However, the referenced tool claims to retrieve the user's open todos while executing an unrestricted db.select().from(todos), potentially returning other users' records and unnecessary fields to hosted inference if the database does not independently enforce isolation.

Content

No source excerpt is available for this finding.