Context Redundancy Deduplicator

Context Redundancy Deduplicator MCP Connector for Claude

A+

Identify overlapping N-grams across RAG documents to calculate redundancy and byte-size savings.

3 tools Official Updated Oct 1, 2026 Official Vinkius Partner

This MCP server provides tools to identify exact overlapping N-gram sequences across multiple retrieved RAG documents. It allows users to compute redundancy percentages, flag documents with high overlap, and calculate exact byte-size savings achievable by removing duplicate text blocks. Use analyze_redundancy to get a full report, find_duplicate_segments to see the specific text blocks, or calculate_savings_projection to estimate storage impact.

ragdeduplicationngramscontext-optimizationdata-analysis

3 tools expose this connector's capabilities to your AI agent.

find_duplicate_segments

Isolate and identify the specific text blocks that are identical across the provided documents

analyze_redundancy

Perform a comprehensive redundancy analysis across a set of documents using a specific N-gram size

calculate_savings_projection

Estimate the impact of deduplication on context window limits or storage

See how to talk to your AI agent using Context Redundancy Deduplicator.

Analyze the redundancy in these documents: [{'documentId': 'doc1', 'content': 'The quick brown fox jumps over the lazy dog. The quick brown fox.'}, {'documentId': 'doc2', 'content': 'The quick brown fox jumps over the lazy dog.'}] using a 5-gram size.

The redundancy analysis shows a redundancy percentage of 45% and identifies that 'doc2' is highly redundant with 'doc1'.

Show me the specific duplicate text segments in these documents using a 10-gram size.

The duplicate segment found is 'the quick brown fox' which appears 3 times.

If I have 10000 bytes of data and a 20% redundancy, how much space can I save?

You can save 2000 bytes, resulting in an estimated new total of 8000 bytes.

An N-gram is a contiguous sequence of N items (characters or words) from a given sample of text used to identify repetitions.

Related Connectors