docs(TSP-1281): add Thread Reply Method to Microsoft Teams trigger docs

[claude]

Summary

• Adds a new Advanced trigger settings section to the Microsoft Teams integration page, positioned after the "How triggered conversations work" subsection
• Documents the Thread Reply Method setting (introduced in PR #15268), which controls whether agents post responses back to Teams after processing a message
• Two options: Agent auto-reply (default, backward compatible) and No reply (silent processing)
• Adds a matching FAQ entry: "Can my agent process Teams messages without replying?"

The new section mirrors the equivalent "No Agent Reply" feature already documented in the Slack integration page.

Linear: https://linear.app/relevance/issue/TSP-1281/

Test plan

• Verify the new section renders correctly in Mintlify preview
• Confirm all headings are in sentence case
• Check the Warning callout displays correctly
• Verify the FAQ accordion opens and reads clearly
• Confirm no broken internal links introduced

🤖 Generated with Claude Code

[linear]

TSP-1281

[mintlify]

Preview deployment for your docs. Learn more about Mintlify Previews.

Project	Status	Preview	Updated (UTC)
relevanceai	🟢 Ready	View Preview	Jun 11, 2026, 3:09 AM

💡 Tip: Enable Workflows to automatically generate PRs for you.

[github-actions]

🎯 Vibe check

Reviewed: 1 file (1 with issues, 0 clean)

Scores

	Dimension	Score	What's holding it back
🟡	Consistency	6/10	"agent/agents" appears lowercase ~15 times when it should be capitalized as the Relevance AI product feature; "workforces" lowercase twice when referring to the product.
🟢	Technical clarity	9/10	Strong — specific UI names, good edge-case documentation, excellent permissions table. "Adaptive Cards" is the only term dropped without a gloss.
🟡	Non-technical clarity	8/10	Solid intro and prerequisites. "Adaptive Cards" is Microsoft-specific jargon that non-technical readers won't recognise.
🟡	Structure	7/10	Good logical flow. Three CardGroup usages where simpler components fit better. Step 3 in the notifications setup is a run-on paragraph covering five distinct choices.

Score key: 🟢 9–10, 🟡 6–8, 🔴 1–5.

✨ Overall vibe: This is a thorough, well-organised integration page — specific UI references, good gotchas (30-min thread expiry, app-per-channel requirement, 401 error diagnosis), and the permissions scopes table is a genuine reader win. The main drag is pervasive lowercase "agent" that needs a global find-replace, plus a handful of CardGroup usages where a bullet list would do the job without the visual overhead.

🔧 Issues (4)

• integrations/popular-integrations/microsoft-teams.mdx:134,136,150,167,172,185,199,224,261,290,394 — "agent/agents" is lowercase throughout when referring to the Relevance AI product. Per CLAUDE.md, Agent is a proper noun. Affected lines include step titles ("Open your agent's settings" line 149), body copy ("Triggers let your agents respond" line 134, "so your agent asks before sending" line 172), and the troubleshooting accordion titles ("Teams trigger option not visible in agent settings", "Can my agent process Teams messages without replying?", "Can my agent access files shared in Teams?"). A global find-replace your agent → your Agent and your agents → your Agents across the file will catch most instances.

• integrations/popular-integrations/microsoft-teams.mdx:136,194 — "workforces" is lowercase when it refers to the Relevance AI Workforce product. Line 136: [workforces](/build/workforces/create-a-workforce) → [Workforces](/build/workforces/create-a-workforce). Line 194: "trigger an entire workforce, not just an individual agent" → "trigger an entire Workforce, not just an individual Agent".

• integrations/popular-integrations/microsoft-teams.mdx:183 — "Adaptive Cards" is dropped without any gloss. Non-technical readers won't know this is a Microsoft message format. Add a short parenthetical: "Responses are delivered as Microsoft Adaptive Cards — rich, interactive message blocks inside Teams."

• integrations/popular-integrations/microsoft-teams.mdx:282 — Step 3 in "Setting up Teams notifications" is one dense paragraph covering five separate choices (platform, trigger type, status selection, account, channel). It's hard to follow on a first read. Break into a short bulleted sub-list within the step, e.g.: select Microsoft Teams → choose notification trigger (status or tool errors) → select statuses if applicable → pick account → pick channel.

🧩 Component suggestions (3)

• integrations/popular-integrations/microsoft-teams.mdx:15–28 — The four prerequisites are in a <CardGroup>, but prerequisites are a checklist, not navigable parallel choices. Cards without links add visual weight without benefit here. Replace with a plain bulleted list (or <Check> items if you want the green ticks):

  - An active project in Relevance AI
  - Desktop or web Microsoft Teams access within your organisation
  - Admin consent if required (see [Admin consent & permissions](#admin-consent-permissions))
  - At least one channel or chat where your Agent should operate
  

• integrations/popular-integrations/microsoft-teams.mdx:34–41 — The "Setup overview" uses a two-card <CardGroup> for sequential steps (Step 1 then Step 2). CardGroup implies parallel choices; these are not interchangeable — both must be done in order. Replace with a two-sentence prose paragraph or a simple numbered list: "Setting up Microsoft Teams requires two steps in sequence: (1) connect your Microsoft account via OAuth, then (2) install the Relevance AI app inside Teams." The detailed <Steps> sections that follow already handle the full instruction, so this overview just needs to set expectations clearly.

• integrations/popular-integrations/microsoft-teams.mdx:296–306 — The "What you'll receive" section uses a <CardGroup cols={2}> with three descriptor cards (Agent status, Direct link, Error details). These are content descriptors, not navigable options. A short bullet list with bold labels is simpler and renders cleanly:

  - **Agent name, status, and task details**
  - **Direct link** to view the full task in Relevance AI
  - **Error details** (for tool errors: which tool failed and why)
  

⚠️ Contradictions (0)

No contradictions found between the Teams page and the Outlook or Slack context pages. The "No reply" trigger behaviour described on this page (lines 207–216) is consistent with the equivalent "No Agent Reply" feature on the Slack page. The unified Microsoft authentication claim (line 9) is not contradicted by the Outlook page.

🔋 Credit usage

Item	Count
Files reviewed	1
Context pages read	2
Total lines processed	~1,121

Files read: integrations/popular-integrations/microsoft-teams.mdx (438 lines), integrations/popular-integrations/outlook.mdx (423 lines), integrations/popular-integrations/slack.mdx (260 lines)