claude-code - 💡(How to fix) Fix MCP M365 Teams chat_message_search timing out (no response)

Official PRs (…)
ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

Helpful · Quick feedback

Loading…

Error Message

  1. Add a default timeout with structured error response (instead of hanging)
RAW_BUFFERClick to expand / collapse

Problem

The Microsoft 365 MCP server's chat_message_search tool times out without returning results, making Teams chat search effectively unusable from Claude Code. This blocks a real workflow — pulling context from team channels before acting on code.

Repro

Tool: mcp__claude_ai_Microsoft_365__chat_message_search

Tried three queries in succession, each with progressively narrower scope to rule out result-set-too-large:

  1. query: "Goldman repo Sextant Dom Spencer frontend", afterDateTime: 2026-05-10, limit: 25TIMED OUT
  2. query: "Sextant Goldman", afterDateTime: 2026-05-12, limit: 15TIMED OUT
  3. query: "main repo", afterDateTime: 2026-05-15, limit: 10TIMED OUT

A fourth query (query: "Ari Devs repo", limit: 10) returned successfully with no output / empty result — which is also worth flagging, since empty results should at minimum return a structured "no matches" response.

Expected behavior

  • A search with reasonable scope (< 25 messages, <30 days lookback) should return within ~10 seconds, or
  • If the search legitimately requires longer, the tool should stream progress / return a partial result handle, not just hang.
  • Empty results should return an explicit "no messages matched" payload, not silent nothing.

Actual behavior

Three consecutive timeout errors with no diagnostic information. No retry guidance, no rate-limit indicator, no way to know if the tenant connection is healthy.

Impact

For users on M365 tenants who need to pull context from Teams (engineering channel discussions, customer threads, internal decisions) before reasoning over code, the connector is currently unreliable. This is a high-value workflow that other MCPs (Notion, Tavily, Gmail) handle without issue.

Asks

  1. Investigate why chat_message_search is hanging — likely Graph API timeout or pagination issue
  2. Add a default timeout with structured error response (instead of hanging)
  3. Return explicit empty-result payload when no matches found
  4. Consider a verboseLogging parameter or telemetry stream to debug from the user side

Environment

  • Claude Code (Opus 4.7, 1M context)
  • macOS Darwin 25.5.0
  • M365 MCP via Claude.ai connector
  • Tenant: Areté Capital Partners
  • User has confirmed authentication (Outlook + SharePoint searches in other sessions work)

🤖 Filed via Claude Code on behalf of Justino @ Areté Capital Partners

Vote matrix · Quick signals

Works
Did the solution work? Tap to confirm.
Easy Fix
Was it a quick fix?
Time Saver
Did it save you time?
Blocking
Was it severely blocking?
Common Issue
Are others likely hitting this too?
Flaky / Intermittent
Is it intermittent?
Verified / Reproducible
Can you reproduce it reliably?
Loading…

FAQ

Expected behavior

  • A search with reasonable scope (< 25 messages, <30 days lookback) should return within ~10 seconds, or
  • If the search legitimately requires longer, the tool should stream progress / return a partial result handle, not just hang.
  • Empty results should return an explicit "no messages matched" payload, not silent nothing.

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

Back to top recommendations

TRENDING

claude-code - 💡(How to fix) Fix MCP M365 Teams chat_message_search timing out (no response)