DocSift

{
  "agent_id": "docsift",
  "name": "DocSift",
  "url": "https://github.com/anishmoncivarghese/docsift",
  "logo": "https://github.com/anishmoncivarghese.png?size=200",
  "niche": null,
  "category": null,
  "short_summary": "Local document processing and retrieval tool that converts PDFs and Office files to Markdown, chunks, and searchable indexes for AI agents via CLI, HTTP API, a…",
  "task_performed": "unclear",
  "inputs_accepted": [],
  "outputs_produced": [],
  "integrations_available": [],
  "protocols_or_interfaces": [],
  "industry_fit": [],
  "autonomy_level": "unclear",
  "human_approval_needed": "unclear",
  "pricing_model": "unclear",
  "price": {
    "observed": false,
    "billing": "not_found",
    "currency": null,
    "lowest_monthly_usd": null,
    "monthly_usd": null,
    "headline": "No public price found",
    "summary": "No public pricing or paid plan amounts were observed; DocSift is presented as a self-hosted open-source project.",
    "confidence": "medium",
    "source_url": "https://github.com/anishmoncivarghese/docsift",
    "checked_at": "2026-10-08T03:24:15.585Z",
    "source_kind": "vendor",
    "amount": null,
    "display": null,
    "plans": [],
    "source": "render+llm"
  },
  "trust_or_rating_signal": [],
  "evidence_quality": "unclear",
  "entity_type": "mcp_server",
  "regulated_data_suitability": "unclear",
  "evidence_urls": [
    "https://github.com/anishmoncivarghese/docsift"
  ],
  "last_checked": null,
  "how_to_connect": {
    "website": "https://github.com/anishmoncivarghese/docsift",
    "docs": null,
    "mcp": null,
    "a2a": null,
    "api": null,
    "protocols": [],
    "install": "pip install \"docsift[markitdown]\" pip install \"docsift[docling]\" pip in",
    "repository": "https://github.com/anishmoncivarghese/docsift"
  },
  "deployment": {
    "model": "both",
    "both_offered": true,
    "hosted_endpoint": null,
    "install": "pip install \"docsift[markitdown]\" pip install \"docsift[docling]\" pip in",
    "note": "Offered both ways: a vendor-hosted endpoint AND a self-run install are documented.",
    "evidence": {
      "sources": [
        "page_text:hosted",
        "page_text:install",
        "page_text:licence",
        "page_text:both_offered"
      ],
      "excerpt": "eing clear about before you install anything. Their connector features only accept a remote MCP server at a public HTTPS URL, and DocSift has no remote transport — there",
      "url": "https://github.com/anishmoncivarghese/docsift",
      "observed_day": "2026-10-07",
      "age_days": 2,
      "hosted_evidence": "wording: \"remote MCP server\"",
      "rule": "deploy-v1"
    }
  },
  "licence": {
    "name": "MIT",
    "note": "Named licence as published; applies to the code, not to where it runs.",
    "evidence": "page text: \"m markdown mcp pdf rag Resources Readme MIT license Contributing Contributing Se\"",
    "observed_day": "2026-10-07",
    "age_days": 2
  },
  "evidence_checks": {
    "price": {
      "status": "unknown",
      "note": "no observed price page text held for this listing"
    },
    "agent_card": {
      "status": "unknown",
      "note": "no Agent Card URL recorded for this listing"
    },
    "endpoint": {
      "status": "unknown",
      "note": "endpoint not yet checked"
    }
  },
  "product_group": null,
  "liveness": {
    "probed": true,
    "alive": false,
    "endpoint_kind": "site",
    "latency_ms": null,
    "uptime_7d": 0.86,
    "checked_at": "2026-10-09T03:28:42.822Z",
    "consecutive_failures": 1,
    "status": "flaky"
  },
  "price_extras": {
    "free_tier": null,
    "unit_cost": null
  },
  "reported_success": null,
  "feedback": "If you actually use this provider, call report_outcome with the result. Testing Agentery's connection or retrieval is not provider use and is stored unweighted."
}