Skip to content
MCP server

FreshContext

Context integrity checks for AI agents: freshness, provenance and confidence

Pricing
Contact sales
Platforms
API, CLI
Runs
Locally or Remotely
Version
0.5.3
Listing updated
1 day ago

About FreshContext

FreshContext is an MCP server for ai models and llm tooling. It can run locally on your machine or as a hosted remote server.

FreshContext addresses a specific failure in retrieval pipelines: a document can match a query semantically while being stale, or belonging to the wrong product, plan, version or region. The README opens with the author asking an assistant for job openings and finding two had vanished and one had been closed for two years.

Its temporal core applies decay-adjusted relevancy, where a source-specific decay rate reduces the base relevance over time, with example half-lives of about 14 hours for Hacker News, 29 days for blogs and 1.6 years for academic papers. Responses are wrapped in an envelope giving source, published date, retrieval time and confidence, defined in an open specification published under MIT.

The npm package is no longer developed, and 0.5.3 is its final release, though it stays available under MIT. The vendor now sells fixed-scope, human-reviewed engagements: a US$25,000 single-workflow assessment, a US$50,000 two-workflow assessment, integration from US$125,000 and enterprise programs from US$250,000.

Install

Run in a terminal

claude mcp add --transport http freshcontext https://api.freshcontext.dev/mcp

Add to Settings → Connectors → Add custom connector

https://api.freshcontext.dev/mcp
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "freshcontext": {
      "url": "https://api.freshcontext.dev/mcp"
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "freshcontext": {
      "type": "http",
      "url": "https://api.freshcontext.dev/mcp"
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "freshcontext": {
      "serverUrl": "https://api.freshcontext.dev/mcp"
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "freshcontext": {
      "type": "http",
      "url": "https://api.freshcontext.dev/mcp"
    }
  }
}

What it does

Decay-adjusted ranking
Lowers the relevance of older context by source type.
Freshness envelope
Wraps content with source, dates and confidence.
Provenance and confidence
Evaluates attribution and utility before context reaches a model.
Signed verdicts
Records verdicts in a verifiable ledger.
Open specification
Publishes the envelope format under MIT.
Assessments
Offers fixed-scope, human-reviewed workflow assessments.

Good for

  • Check that retrieved documents are still current
  • Rank RAG results with time decay
  • Audit an agent workflow for stale or mis-scoped context
  • Record why a piece of context was accepted or held

Pricing

Contact sales

Fixed-scope engagements from US$25,000; MIT npm package no longer developed

Prices change. Check freshcontext.dev before you buy.

FreshContext FAQ

Is the package still maintained?

No. Version 0.5.3 is the last release; it stays available under MIT.

What does the vendor sell now?

Fixed-scope engagements, starting at a US$25,000 single-workflow assessment.

Does it determine truth?

No. The vendor says it evaluates supported context under explicit scope and policy, not truth.

Is it fully autonomous?

No. The commercial assessment is human-reviewed.

How much does FreshContext cost?

FreshContext does not publish prices. Contact the company for a quote.

How do I add FreshContext to Claude Code?

Run this command in your terminal: claude mcp add --transport http freshcontext https://api.freshcontext.dev/mcp

What are the best alternatives to FreshContext?

The closest alternatives to FreshContext are Codebase Memory, XERJ, Claude Code Ultimate Guide, Nautilus Compass Memory and note1.

Alternatives to FreshContext

  1. Codebase Memory

    MCP server

    Local MCP server that builds a code knowledge graph for AI coding agents

    Runs locally
    • Open source
    • 48.7K GitHub stars
    • Knowledge and memory
  2. XERJ

    MCP server

    Local search engine for AI agents with cited answers over your code and docs

    Runs locally
    • Open source
    • 19.2K GitHub stars
    • Knowledge and memory
  3. MCP server for searching the Claude Code Ultimate Guide from any MCP client

    Runs locally
    • Open source
    • 6.5K GitHub stars
    • AI models and LLM tooling
  4. Local-first long-term memory layer for AI agents, served over MCP

    Not checked yet
    • 5 tools
    • Remote
    • Open source
    • 1.3K GitHub stars
    • Knowledge and memory
  5. note1

    MCP server

    MCP server giving AI assistants search and summaries across your note1 meetings

    • 4 tools
    • Remote
    • Free trial
    • 1 GitHub star
    • Productivity