Book a 30-minute demo →
Skill scan report

web-research-assistant

View on GitHub
20 Low This skill sends sensitive values to an external host. Automated analysis flagged 1 additional risk pattern.

What this skill does

AI-powered web research assistant that uses BrowserAct API to search the internet for information when primary web access is restricted or blocked

github/browser-act - Data Exfiltration - 5.3k stars

ThreatsData Exfiltration Secrets Management

Threat analysis

Data Exfiltration1 finding
Secrets Management1 finding

Skill info

Namebrowser-act/web-research-assistant
Registrygithub
Version4577dc5
PURLpkg:github/browser-act/skills@4577dc5?skill=web-research-assistant
Stars5.3k

Assessments (2)

Data Exfiltration1 finding MEDIUM
MEDIUM

Data Exfiltration via local-llm-review

scripts/research.py

The script sends search results to MCP_SERVER_URL (https://mcp.browseract.com/) using the BROWSERACT_MCP_TOKEN for authorization. This could potentially exfiltrate search results or other data to a th
Secrets Management1 finding MEDIUM
MEDIUM

Secrets Management via local-llm-review

scripts/research.py

The script uses environment variables BROWSERACT_API_KEY and BROWSERACT_MCP_TOKEN, which could be at risk if not properly secured or managed.

Badge

Add the Anomity scan badge for web-research-assistant to your README.

Anomity Skill Check badge

Markdown
[![Anomity Skill Check](https://anomity.ai/skills/badge.svg)](https://anomity.ai/skills/github/browser-act/web-research-assistant/)
HTML
<a href="https://anomity.ai/skills/github/browser-act/web-research-assistant/"><img src="https://anomity.ai/skills/badge.svg" alt="Anomity Skill Check"></a>
Image URL
https://anomity.ai/skills/badge.svg

How Anomity governs this at runtime

Scan-time vetting tells you what a skill says it will do. Anomity's Endpoint Sensor sees what agents actually do: it discovers skills alongside every other AI artifact on the endpoint, and runtime governance can allow, deny, or log the tool calls a skill triggers. Policy violations route to your SIEM, Slack, email, or Jira, backed by a queryable 90-day audit trail.

Book a 30-minute demo to see your own skill inventory.

Methodology and disputes

Every skill is assessed by the Anomity Skill Intelligence engine against its public source; findings indicate risk patterns, not confirmed exploitation. Maintainer of web-research-assistant? Report an issue or request a rescan.

Ask AI about Anomity
ChatGPT Claude Perplexity Google AI Grok