
Visus Mcp
Built for production AI agents that need to fetch web content without eating prompt injection or leaking PII. Runs a Playwright-based pipeline that detects 19 categories of indirect prompt injection before sanitization, redacts emails and credit cards, and generates cryptographic proofs that the security pass actually ran. Handles PDF, JSON, and SVG in addition to HTML. The threat detection is granular: you get severity scores, confidence levels, and character offsets for each flagged attack. Designed around EU AI Act compliance requirements, with audit logs and Merkle-chained session ledgers. If your agent reads untrusted web pages and you need both safety and regulatory coverage, this is the tooling for it.
- GitHub stars
- 1
- tools
- 10
- transport
- STDIO
- env vars
- 3
- last updated
- Apr 2, 2026
Tools
10 toolsPublic tool metadata for what this MCP can expose to an agent.
visus_fetchFetch and sanitize a web page with automatic format detection, supporting HTML, JSON, XML, and RSS/Atom feeds.
Fetch and sanitize a web page with automatic format detection, supporting HTML, JSON, XML, and RSS/Atom feeds.
No parameter schema in public metadata yet.
visus_readExtract clean article content from a web page using Mozilla Readability (reader mode).
Extract clean article content from a web page using Mozilla Readability (reader mode).
No parameter schema in public metadata yet.
visus_searchSearch the web via DuckDuckGo and return sanitized results with prompt injection and PII removed.
Search the web via DuckDuckGo and return sanitized results with prompt injection and PII removed.
No parameter schema in public metadata yet.
visus_fetch_structuredExtract structured data from a web page according to a schema.
Extract structured data from a web page according to a schema.
No parameter schema in public metadata yet.
visus_verifyVerify a Visus-MCP sanitization proof record, confirming a request was processed by the injection detection pipeline.
Verify a Visus-MCP sanitization proof record, confirming a request was processed by the injection detection pipeline.
No parameter schema in public metadata yet.
visus_get_ledger_proofRetrieve tamper-evident proof for a specific request ID, including event details and Merkle inclusion proof.
Retrieve tamper-evident proof for a specific request ID, including event details and Merkle inclusion proof.
No parameter schema in public metadata yet.
visus_read_csvReads and sanitizes a CSV or TSV file from a local path or URL.
Reads and sanitizes a CSV or TSV file from a local path or URL.
No parameter schema in public metadata yet.
visus_read_excelReads and sanitizes an Excel workbook from a local path or URL.
Reads and sanitizes an Excel workbook from a local path or URL.
No parameter schema in public metadata yet.
visus_read_gsheetReads and sanitizes a public Google Sheet.
Reads and sanitizes a public Google Sheet.
No parameter schema in public metadata yet.
visus_context_scanDetect multi-turn priming risks in conversation history.
Detect multi-turn priming risks in conversation history.
No parameter schema in public metadata yet.
Sponsors
Configuration
VISUS_TIMEOUT_MSOptional timeout in milliseconds for page fetching (default: 10000)
VISUS_MAX_CONTENT_KBOptional maximum content size in KB before truncation (default: 512)
VISUS_RENDERER_URLOptional URL for managed Lambda renderer (Phase 2 - not yet available)
Readme
Details
- Repository
- github.com/visus-mcp/visus-mcp
- Website
- github.com/visus-mcp/visus-mcp
- Categories
- Design & Creative, Documents & Knowledge, Security & Pentesting, Data & Analytics
- Package
visus-mcp- Transport
- STDIO
- Registry
- MCP registry, active