Skip to content
agent-browser-shield

agent-browser-shield

Prototype browser extension capabilities that make browser-use agents faster, safer, and more accurate.

Install from Chrome Web StoreOther install pathsBrowse the rulesView on GitHubSkill on ClawHub
Before agent-browser-shield
Before
After agent-browser-shield
After

Benefits

Section titled “Benefits”

Token efficiency

Strip page chrome — footers, cookie banners, chat widgets, sponsored content — so agents spend tokens on the user’s task.

Security & compliance

Mask PII and credentials before they reach the model, and suppress hidden text, HTML comments, and user-generated content that could carry prompt-injection payloads.

Accuracy

Block manipulative dark patterns and hide engagement rails and other content that could distract the model from the user’s task.

Ideas in the repo

Section titled “Ideas in the repo”
  • Masking and redacting sensitive information on the webpage.
  • Blocking and modifying dark patterns on the webpage.
  • Preprocessing webpage content to be more agent-friendly — hiding irrelevant content, hiding user-generated comments that could contain prompt-injection attacks, and similar.

For the full project overview, see the repository README.

Privacy

Section titled “Privacy”

The extension does not collect, store, or send any telemetry, analytics, or usage data. Rule processing runs locally in your browser; nothing is reported back to PixieBrix or any other server.

The one outbound network call the extension can make is the optional irrelevant-sections-redact rule (off by default), which sends a compressed page tree to OpenAI’s API for classification when you enable the rule and configure an API key.