Add documentation for Inventor scheduled check-ins management

[claude]

Summary

• Adds a new Scheduled check-ins tab to the Inventor features section in build/invent/invent.mdx, documenting the clock icon, the 200-item panel, reschedule/cancel actions, and early-access status
• Adds a brief cross-reference in build/agents/build-your-agent/agent-triggers/scheduled-triggers.mdx pointing users to the Inventor panel for centralized check-in management

Documents the new scheduled check-ins management feature added in PR #16006. Feature is currently in early access.

Linear: https://linear.app/relevance/issue/TSP-1337/

Test plan

• Verify the new "Scheduled check-ins" tab renders correctly in the Inventor features <Tabs> block
• Confirm the <Info> callout in the tab and in scheduled-triggers.mdx renders without attributes (correct Mintlify syntax)
• Confirm the cross-reference link /build/invent/invent resolves correctly
• Check all headings are in sentence case
• Verify no banned words (powerful, seamlessly, etc.) were introduced

🤖 Generated with Claude Code

[linear]

TSP-1337

[mintlify]

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

Project	Status	Preview	Updated (UTC)
relevanceai	🟢 Ready	View Preview	Jul 10, 2026, 4:59 AM

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

[github-actions]

🎯 Vibe check

Reviewed: 2 files (2 with issues, 0 clean)

Scores

	Dimension	Score	What's holding it back
🟡	Consistency	7/10	British spelling (behaviour), title-case bold labels acting as headings, and lowercase product feature names (Agent, Tool, etc.) scattered across both files.
🟡	Technical clarity	8/10	Steps in scheduled-triggers.mdx use single quotes for UI element names instead of bold; model list in invent.mdx described as "the configurable list" with no specifics.
🟢	Non-technical clarity	9/10	Both pages define their concepts before instructions. Inventor page is well-structured for a concept page.
🟡	Structure	7/10	invent.mdx is a concept page with no closing CTA. Two CardGroups in invent.mdx use cards for non-navigable property descriptors where bullet lists would be lighter.

Score key: 🟢 9–10, 🟡 6–8, 🔴 1–5.

✨ Overall vibe: Both pages are clear and well-organized with a good logical flow — the Inventor page in particular is a solid concept page. The main drag is a handful of consistency slips (one British spelling, product feature terms inconsistently capitalized, labels using title case) and two CardGroups that are doing work bullets would do better.

🔧 Issues (10)

• build/agents/build-your-agent/agent-triggers/scheduled-triggers.mdx:114 — behaviour → behavior (British spelling; American English required per CLAUDE.md)

• build/agents/build-your-agent/agent-triggers/scheduled-triggers.mdx:45 — **Cron Expression Format:** → **Cron expression format:** (title case in a bold label acting as a section header)

• build/agents/build-your-agent/agent-triggers/scheduled-triggers.mdx:67 — **Common Cron Expression Examples:** → **Common cron expression examples:** (same issue)

• build/agents/build-your-agent/agent-triggers/scheduled-triggers.mdx:19–24 — Steps use single quotes around UI element names ('Edit Agent', 'Triggers', 'Recurring Schedule'). CLAUDE.md requires bold: Click **Edit Agent**, Navigate to **Triggers**, etc.

• build/agents/build-your-agent/agent-triggers/scheduled-triggers.mdx:20 — "Go to your agent in the Relevance AI dashboard" — "agent" should be "Agent" (referring to the Relevance AI product feature, not the generic concept)

• build/agents/build-your-agent/agent-triggers/scheduled-triggers.mdx:110 — "your agent's core instructions" → "your Agent's core instructions"

• build/agents/build-your-agent/agent-triggers/scheduled-triggers.mdx:114 — "your agent will understand time-based requests" → "your Agent will understand" (same product-feature capitalization rule)

• build/invent/invent.mdx:4 — Frontmatter description: "agents, tools, workforces, and triggers" → "Agents, Tools, Workforces, and Triggers" (these are Relevance AI product feature names throughout the rest of the page)

• build/invent/invent.mdx:137 — "agents, tools, workforces, and triggers" → "Agents, Tools, Workforces, and Triggers" (same rule; the list appears in Step 2 under "How to use Inventor")

• build/invent/invent.mdx:232 — "letting external clients call your tools, and letting your agents reach into external services" → "your Tools" and "your Agents" (product feature references)

🧩 Component suggestions (2)

• build/invent/invent.mdx:172–182 — The "About you" tab uses a 3-column <CardGroup> (Tone and language / Domain assumptions / Personalization) to describe three properties of what "About you" changes. These aren't navigable items or parallel choices — they're short descriptors. Per CLAUDE.md, CardGroup is not appropriate for property lists like this. A simple bullet list (- **Tone and language:** Simpler explanations for non-technical users…) carries the same information with less visual overhead.

• build/invent/invent.mdx:205–212 — The "Feedback" tab uses a 2-item <CardGroup> for thumbs up / thumbs down. Two cards with icons for "The response was accurate" and "The response was incorrect" is heavier than needed — this is a 2-line property list. Prose or two bullets work fine here.

🏗️ Page structure (1)

• build/invent/invent.mdx — The page ends with a <Tip> that mentions the MCP, but no closing CTA. This is a concept page (explaining what Inventor is), so a "What's next?" section is warranted. Natural next steps to link would be /build/agents/create-an-agent (the most common first action after learning about Inventor) and /integrations/mcp/programmatic-gtm/mcp-server (already mentioned in the MCP comparison section). A third option might be /build/agents/build-your-agent/evals since it's called out in the "What you can do" cards.

⚠️ Contradictions (0)

No contradictions found. The <Info> in scheduled-triggers.mdx linking to the scheduled check-ins panel in Inventor (/build/invent/invent) is consistent with the "Scheduled check-ins" tab content in invent.mdx.

🔋 Credit usage

Item	Count
Files reviewed	2
Context pages read	1
Total lines processed	~536

Files read: build/agents/build-your-agent/agent-triggers/scheduled-triggers.mdx (163 lines), build/invent/invent.mdx (250 lines), build/agents/build-your-agent/agent-triggers/api-trigger.mdx (123 lines, sibling context)