Exa — Neural Web Search & Research

Name: web-search-exa
Author: theishangoswami

Share on X Share on LinkedIn Share on Bluesky

40Medium

The web-search skill is vulnerable to SSRF and exposes API keys in URL parameters, risking internal network access and credential compromise.

ThreatsCI SM CS BC S HITL

Behavior Analysis

Description verified — behavior matches claim

Claims to do

Exa — Neural Web Search & Research: Exa is a neural search engine. Unlike keyword-based search, it understands meaning — you describe the page you're looking for and it finds it. Returns clean, LLM-ready content with no scraping needed.

Actually does

This skill interacts with the Exa MCP server at `https://mcp.exa.ai/mcp` to perform neural web searches, extract clean content from specified URLs, and conduct specialized research (e.g., company, people, code context). It uses various tools like `web_search_exa`, `crawling_exa`, and `deep_search_exa`, optionally leveraging an API key for advanced features and higher rate limits.

Threat Analysis

AI Agent Traps ↗Scanned 4/15/2026

Content InjectionPerceptionclean

Semantic ManipulationReasoningclean

Cognitive StateMemory & Learningclean

Behavioural ControlAction2 findings

SystemicMulti-Agent Dynamicsclean

Human-in-the-LoopHuman Overseerclean

Skill Info

Nametheishangoswami/web-search-exa

Registryclawhub

Version2.0.0

PURLpkg:clawhub/theishangoswami/web-search-exa@2.0.0

Stars44

Downloads25,830

Installs54

Source

Repository ↗

SHA-25624f08ac90b30...

SHA-1bba1c39b92bd...

MD5eaddf7a96e3d...

TLSHe232f9766c7b...

ssdeep192:DWzvGMu0...

Size11,769 bytes

Install

This skill has been flagged as potentially malicious. Review the findings below before installing.

OpenClawDocs

https://clawhub.ai/theishangoswami/web-search-exa

25,830 downloads54 installs

ClawHubDocs

openclaw skills install theishangoswami/web-search-exa

Assessments (1)

AI Agent Traps ↗

Arbitrary URL crawling capability (SSRF potential)mediumssrf

The `crawling_exa` tool allows fetching content from any specified URL. This capability could be abused by a malicious agent to perform Server-Side Request Forgery (SSRF) against the backend service, potentially probing internal network resources or cloud metadata endpoints if the service's URL validation is insufficient.

80% confidenceline 254

crawling_exa { "url": "https://arxiv.org/abs/2301.07041" }

API key exposure risk in URL parameterslowcredential theft

The skill instructs users to pass the API key as a URL query parameter. This method can increase the risk of the key being exposed in server logs, browser history, or network traffic compared to using HTTP headers.

90% confidenceline 38

https://mcp.exa.ai/mcp?exaApiKey=YOUR_EXA_KEY

Badge

Markdown

[![Mondoo Skill Check](https://mondoo.com/ai-agent-security/api/badge/clawhub/theishangoswami/web-search-exa.svg)](https://mondoo.com/ai-agent-security/skills/clawhub/theishangoswami/web-search-exa)

HTML

<a href="https://mondoo.com/ai-agent-security/skills/clawhub/theishangoswami/web-search-exa"><img src="https://mondoo.com/ai-agent-security/api/badge/clawhub/theishangoswami/web-search-exa.svg" alt="Mondoo Skill Check" /></a>

Image URL

https://mondoo.com/ai-agent-security/api/badge/clawhub/theishangoswami/web-search-exa.svg

Secure your AI agents

Skills can read files, run commands, and access credentials. Mondoo helps organizations manage the security risks of AI agent skills across their entire fleet.

Continuous scanning of skills across all registries
Policy enforcement before skills reach your agents
Integration with your existing security workflow