How to search email archives and attachments with an AI agent, locally Leave feedback

The GroupDocs.Search MCP server opens email files and archives — PST, OST, MSG, EML — and indexes each message and its attachments, locally in Docker. Ask the agent a question, and a match inside an attached document comes back with the message file and the attachment’s name:

docker run --rm -i -v $(pwd)/documents:/data \
  ghcr.io/groupdocs-search/search-net-mcp:latest

Index the mail folder. Which message mentions a price increase, and what exactly does it say?

Note

The commands and config snippets on this page are for the .NET build of the server, which ships as a Docker image only. Installation and client setup: MCP server for .NET. Other platforms will expose the same tools with their own launch command; everything else on this page applies unchanged.

• The archive or email files inside the folder mounted at /data — export nothing, convert nothing.
• For archives larger than a few hundred messages, a license: evaluation mode stops at 100 documents per index, and every message and attachment counts.

mail/2026-09-18-re-q3-pricing.eml          <- filePath
  |- body: "please find the updated quotation attached"
  '- attachment: quotation-Q-2026-118.docx  <- innerPath
        "...we announce a price increase of 6 percent..."

search {"query":"\"price increase\"","indexName":"mail"}
  -> {"filePath":"mail/2026-09-18-re-q3-pricing.eml",
      "innerPath":"quotation-Q-2026-118.docx", "occurrences":1}

Captured from server 26.10.1 with three EML messages; the phrase occurs only inside the Word quotation attached to one of them:

-> add_to_index {"documentsPath":"mail","indexName":"mail"}
<- Indexed 'mail' into index 'mail': 3 document(s) added, 3 in the index.

-> search {"query":"\"price increase\"","indexName":"mail"}
<- { "documentsFound": 1, "documents": [
     { "filePath": "mail/2026-09-18-re-q3-pricing.eml",
       "innerPath": "quotation-Q-2026-118.docx",
       "occurrences": 1 } ] }

-> get_text_fragments {"query":"\"price increase\"","indexName":"mail"}
<- === mail/2026-09-18-re-q3-pricing.eml → quotation-Q-2026-118.docx
   (1 occurrence(s)) ===
     [content] …Due to raw material costs we announce a price increase
   of 6 percent from 2026-11-01.…

-> search {"query":"invoice AND 1042","indexName":"mail"}
<- documentsFound 1: mail/2026-09-22-invoice-1042.eml (6 occurrences)

The message body says only “please find the updated quotation attached”; the answer was in the attachment, and innerPath says which one.

Which message from Fabrikam mentions a delivery delay?

fabrikam AND delay → the delivery-delay message. Sender addresses are indexed with the message: contoso also finds the invoice whose body never names the company.

Find every email that mentions order 77812.

77812 — whole numbers are words.

Which invoices fall due in October?

invoice AND content: daterange(2026-10-01 ~~ 2026-10-31) → invoice 1042, “due on 2026-10-22”. It matches dates written in the message text — the Date: header uses a format the range does not read; see the date-range note.

The message file, the attachment where relevant, and the quoted passage — from an archive that never left the machine and was never converted.

Does it need Outlook installed? No. The engine reads PST, OST, MSG and EML directly.

Is anything uploaded? No. The archive is read inside the local container; only what the agent quotes back reaches your model provider.

Can it search the archive and the documents folder together? Index them into one index, or keep mail and documents as separate indexes and ask the agent to search both.

Agents follow project instructions about when to use a tool. Paste this into the instruction file your agent reads (CLAUDE.md, AGENTS.md, .github/copilot-instructions.md, .cursorrules):

Email under ./documents/mail is indexed in the groupdocs-search index 'mail'.
For questions about correspondence, search that index; report filePath and innerPath
(the attachment) for every match, and quote with get_text_fragments.

• Evaluation mode: an index holds at most 100 documents and a search returns at most 10 — the response says so in its evaluation field. Licensing.
• Full-text, not semantic: it matches words, not meaning. Ask the agent to try variants (OR, fuzzy: true).
• No OCR: scans without a text layer are indexed by their file properties only.
• Docker only: there is no dnx channel for this product (why).

• Search documents with AI agents
• Search a folder of contracts
• Quote matching passages
• Keep an index up to date
• Search documents in a code repository
• Find documents by date and amount
• On-premise deployment
• Tools reference · Licensing · Troubleshooting & FAQ
• Questions: Search forum

• Search your email archive with an AI agent: PST, MSG, EML and the attachments inside — the same scenario on the GroupDocs blog, with worked examples and the trade-offs.