docs: group cross-cutting features under a new Platform section

[jordanc-relevanceai]

What

Adds a Platform section to the Build with Relevance AI tab, alongside Overview, Invent, Agents, Tools, Workforce, and Knowledge.

Evals, Tasks, Sharing, and Version history all work across more than one builder object, but each was filed under a single one. They're now in one place.

Build with Relevance AI
├── Overview
├── Invent
├── Agents
├── Tools
├── Workforce
├── Knowledge
└── Platform          ← new
    ├── Evaluating Performance   (dropdown, from Agents)
    ├── Running Tasks            (dropdown, from Agents)
    ├── Sharing                  (dropdown, new — from Agents, Tools, Workforce)
    └── Version history          (single page, from Agents › Build an Agent)

Dropdowns are kept where they already existed. Version history is one document, so it stays one document.

Moves

Old	New
build/agents/evals/*	build/platform/evals/*
build/agents/give-your-agent-tasks/* (the 5 nav pages)	build/platform/tasks/*
build/agents/share-your-agent	build/platform/sharing/share-your-agent
build/tools/share-your-tool	build/platform/sharing/share-your-tool
build/workforces/share-your-workforce	build/platform/sharing/share-your-workforce
build/agents/build-your-agent/version-history	build/platform/version-history

Also in here

• 14 new redirects entries, one per moved URL, plus three existing /evals redirects repointed at the new Evals introduction
• Inbound links updated across 36 files (guides, changelog, pricing, RBAC, quick start, and the moved pages themselves)
• The three Sharing pages get sidebarTitle values of Agents / Tools / Workforces so the dropdown reads cleanly, and Share Your Agent is corrected to sentence case

No page files were deleted, and no content was rewritten.

mintlify broken-links passes.

Deliberately left alone

delete-agent, delete-tool, and inactive-table-deletion stay where they are — deleting is a per-object action rather than a platform feature, and the Knowledge one is about automatic cleanup, not the same thing. Easy follow-up if you want a Deleting dropdown too.

🤖 Generated with Claude Code

[mintlify]

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

Project	Status	Preview	Updated
relevanceai	🟢 Ready	View Preview	Sep 7, 2026, 12:49 AM

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

[github-actions]

🎯 Vibe check

Reviewed: 51 files (11 with issues, 40 clean)

Scores

	Dimension	Score	What's holding it back
🟡	Consistency	7/10	Heading title case violations in task-queue.mdx, rbac.mdx, interact-with-your-agent.mdx, and test-sets.mdx; product term ("Agent", "Agents", "Tools", "Knowledge") not capitalized in several quick-start-guide.mdx steps; copy-paste duplicate text in share-your-tool.mdx.
🟡	Technical clarity	7/10	interact-with-your-agent.mdx links to three pages that don't exist in the nav tree using absolute URLs; share-your-tool.mdx duplicates the wrong section content under Embed; grammar error in bulk-schedule.mdx; critical contradiction between pricing.mdx (Evals is Enterprise-only) and evals/introduction.mdx ("Evals is available on every plan").
🟡	Non-technical clarity	8/10	interact-with-your-agent.mdx reads like a landing page, not documentation — "users will love", "professional 'storefront'", "Why You'll Love It", "Perfect For". The 27 guide files and the Evals suite are well-written throughout.
🟡	Structure	8/10	interact-with-your-agent.mdx has bold text inside a heading, a frontmatter description that doesn't match the content, and a "Related Features" section that's a dead end. All other pages follow logical structure.

Score key: 🟢 9–10, 🟡 6–8, 🔴 1–5.

✨ Overall vibe: The 27 vertical guide files (sales, marketing, revops, customer support, etc.) are a consistent, high-quality addition — clean templating, proper capitalization, no banned words. The Evals documentation suite is similarly solid. The platform-level files are more mixed: most are good, but interact-with-your-agent.mdx is a significant outlier that reads like marketing copy and has broken links, and there's a factual contradiction between the pricing page and the Evals intro that needs resolving before merge.

🔧 Issues (21)

build/platform/tasks/interact-with-your-agent.mdx

• :1 — sidebarTitle: "Interact With Your Agent" → "Interact with your Agent" (sentence case)
• :4 — description: "What are tasks and how do they differ from prompt?" doesn't describe the page content (which is entirely about Agent Chat). Update to match.
• :7 — ## **Agent Chat** — bold inside a heading; use ## Agent Chat. CLAUDE.md: "Don't mix bold text with heading formats for section titles."
• :13,23,25,32,35,38,48,57,63,71,78 — All H3 headings are in title case: ### Why You'll Love It, ### Key Features, ### Conversational Interface, ### Upload Files, ### Talk to Your Agent, ### Embed Anywhere, ### Customization Options, ### How to Set It Up, ### Important Considerations, ### Perfect For, ### Related Features — all need sentence case.
• :79–84 — ### Related Features links use absolute URLs (https://relevanceai.com/docs/build/agents/templates, .../customise-agent/agent-profile, .../customise-agent/share-agent) instead of root-relative paths. All three target pages (/build/agents/templates, /build/agents/customise-agent/agent-profile, /build/agents/customise-agent/share-agent) don't appear anywhere in docs.json. These are likely dead links to pages that were removed or renamed. The correct share page is /build/platform/sharing/share-your-agent.
• :9–12 — Marketing language: "friendly, modern chat interface that your users will love", "professional 'storefront' where people can easily interact with it" — replace with plain description of what the interface is.
• :13–20 — "Why You'll Love It" section uses bold bullet labels (**Looks Professional**, **Works Anywhere**, **Fully Customizable**, **User-Friendly**) as marketing copy. CLAUDE.md: "Don't mix bold text with heading formats for section titles." Convert to H4 headings or prose.

build/platform/tasks/task-queue.mdx

• :1–5 — sidebarTitle: 'Task Queue' → "Task queue"
• :7,12,26,45,63,74 — All section headings in title case: ## Understanding the Task Queue, ## How to Access the Task Queue, ## Task Statuses, ## Cancelling Tasks, ## Viewing Queue Insights, ## Best Practices — all need sentence case.

enterprise/rbac.mdx

• :22,26,34,45,138 — Headings in title case: ## Best Practices, ### Organization Level, ### Project Level, ### Asset Level, ### Chat Role Details — all need sentence case.
• :237 — <Info>**Note:** If a user has run access... — bold label inside a callout. CLAUDE.md: "Callouts must be a single short paragraph — no bullet lists, no multi-line content, no bold labels inside." Remove the **Note:** prefix.

build/platform/evals/test-sets.mdx

• :13 — ## How to Create a Test → "How to create a Test"

build/platform/sharing/share-your-agent.mdx

• :9 — "your AI agent" → "your AI Agent"
• :17 — "Cloned agents use the recipient's credits" → "Cloned Agents use…"
• :42,60 — #### Developer Notes (appears twice) → "Developer notes" (sentence case)

build/platform/sharing/share-your-tool.mdx

• :9 — Click on `Share` uses backtick code formatting for a UI element. Per CLAUDE.md: "CORRECT: Click **Create agent**" — should be Click on **Share**.
• :59–63 — "When sharing your Tool as a sharable link, you can select:" appears inside the ## Embed your Tool section — verbatim copy of the sharable link intro. Should describe embed-specific customization options, not a sharable link.

build/platform/sharing/share-your-workforce.mdx

• :29–33 — Steps 1–5 use single quotes for UI elements ('Run', 'Share', 'Enable public access', 'Copy') instead of bold formatting. Per CLAUDE.md: "CORRECT: Click **Create agent**" — use **Run**, **Share**, **Enable public access**, **Copy**.

build/platform/tasks/bulk-schedule.mdx

• :75 — "This give you the option" → "This gives you the option" (grammar error)

get-started/quick-start-guide.mdx

• :42,49,50,80,81 — Product terms written lowercase where the parallel file (build/agents/create-an-agent.mdx) uses uppercase for the same steps: "the generated agent" → "Agent"; "suggest tools" → "Tools"; "add [tools]" → "[Tools]"; "add [knowledge]" → "[Knowledge]". Creates inconsistency across the two pages covering the same flows.

🧩 Component suggestions (2)

• build/platform/tasks/interact-with-your-agent.mdx:57–62 — "How to Set It Up" is a numbered sequential procedure (Configure → Customize → Test → Deploy). Wrap in <Steps> for visual progress indicators: <Step title="Configure">In your Relevance AI dashboard, set up your Agent Chat preferences</Step> etc.
• build/platform/tasks/interact-with-your-agent.mdx:13–20 — The "Why You'll Love It" bold-bullet list would be cleaner as a <CardGroup cols={2}> with icons matching the benefit type, or collapsed to prose. The current bold-bullet format is the anti-pattern CLAUDE.md flags: "Don't mix bold text with heading formats for section titles."

🏗️ Page structure (1)

• build/platform/tasks/interact-with-your-agent.mdx — The page doesn't fit a clear type. It opens as a concept page (what is Agent Chat), then pivots to a feature list with marketing framing ("Why You'll Love It", "Perfect For"), then ends with a setup procedure and a Related Features section that links to dead pages. Suggest restructuring as a clean concept page: definition → what it includes → how to set it up → next steps linking to /build/platform/sharing/share-your-agent.

⚠️ Contradictions (1)

• get-started/pricing.mdx:164 marks Agent Evaluations as Enterprise-only (✗ for Free, Pro, and Team). build/platform/evals/introduction.mdx:98 states in the FAQ: "Evals is available on every plan." Both files are in this PR. One of these is wrong — align them before merge. If Evals truly requires Enterprise, update the FAQ; if it's available on all plans, update the feature comparison table.

✅ Clean files (40)

build/agents/create-an-agent.mdx, build/invent/invent.mdx, build/platform/evals/checks.mdx, build/platform/evals/monitor.mdx, build/platform/evals/running-evaluations.mdx, build/platform/tasks/task-overview.mdx, build/platform/tasks/tasks-page.mdx, build/platform/version-history.mdx, build/tools/create-a-tool.mdx, build/workforces/create-a-workforce.mdx, changelog.mdx, get-started/troubleshooting/troubleshooting-agents-not-working.mdx, guides/customer-success/account-health.mdx, guides/customer-success/getting-started.mdx, guides/customer-success/qbr-prep.mdx, guides/customer-success/renewal-expansion.mdx, guides/customer-support/getting-started.mdx, guides/customer-support/kb-generation.mdx, guides/customer-support/response-drafting.mdx, guides/customer-support/ticket-triage.mdx, guides/marketing/campaign-analytics.mdx, guides/marketing/content-repurposing.mdx, guides/marketing/getting-started.mdx, guides/marketing/lifecycle-campaigns.mdx, guides/revops/data-dedup.mdx, guides/revops/getting-started.mdx, guides/revops/lead-routing.mdx, guides/revops/pipeline-hygiene.mdx, guides/sales/competitive-intelligence.mdx, guides/sales/crm-data-enrichment.mdx, guides/sales/getting-started.mdx, guides/sales/lead-scoring.mdx, guides/sales/meeting-briefs.mdx, guides/sales/personalized-outbound.mdx, guides/sales/prospect-research-use-case.mdx, guides/solution-engineering/demo-prep.mdx, guides/solution-engineering/discovery-summaries.mdx, guides/solution-engineering/getting-started.mdx, guides/solution-engineering/rfp-responses.mdx, docs.json (partial read, navigation structure looks sound)

🔋 Credit usage

Item	Count
Files reviewed	51
Context pages read	0
Total lines processed	~5,700

Files read: build/agents/create-an-agent.mdx (92), build/invent/invent.mdx (360), build/platform/evals/checks.mdx (123), build/platform/evals/introduction.mdx (105), build/platform/evals/monitor.mdx (55), build/platform/evals/running-evaluations.mdx (110), build/platform/evals/test-sets.mdx (132), build/platform/sharing/share-your-agent.mdx (118), build/platform/sharing/share-your-tool.mdx (69), build/platform/sharing/share-your-workforce.mdx (39), build/platform/tasks/bulk-schedule.mdx (93), build/platform/tasks/interact-with-your-agent.mdx (84), build/platform/tasks/task-overview.mdx (53), build/platform/tasks/task-queue.mdx (84), build/platform/tasks/tasks-page.mdx (40), build/platform/version-history.mdx (89), build/tools/create-a-tool.mdx (103), build/workforces/create-a-workforce.mdx (46), changelog.mdx (100, partial), enterprise/rbac.mdx (382), get-started/pricing.mdx (271), get-started/quick-start-guide.mdx (137), get-started/troubleshooting/troubleshooting-agents-not-working.mdx (195), docs.json (200, partial), guides/customer-success/account-health.mdx (96), guides/customer-success/getting-started.mdx (187), guides/customer-success/qbr-prep.mdx (96), guides/customer-success/renewal-expansion.mdx (96), guides/customer-support/getting-started.mdx (187), guides/customer-support/kb-generation.mdx (96), guides/customer-support/response-drafting.mdx (96), guides/customer-support/ticket-triage.mdx (96), guides/marketing/campaign-analytics.mdx (96), guides/marketing/content-repurposing.mdx (96), guides/marketing/getting-started.mdx (196), guides/marketing/lifecycle-campaigns.mdx (96), guides/revops/data-dedup.mdx (96), guides/revops/getting-started.mdx (189), guides/revops/lead-routing.mdx (96), guides/revops/pipeline-hygiene.mdx (96), guides/sales/competitive-intelligence.mdx (96), guides/sales/crm-data-enrichment.mdx (96), guides/sales/getting-started.mdx (196), guides/sales/lead-scoring.mdx (96), guides/sales/meeting-briefs.mdx (96), guides/sales/personalized-outbound.mdx (96), guides/sales/prospect-research-use-case.mdx (105), guides/solution-engineering/demo-prep.mdx (96), guides/solution-engineering/discovery-summaries.mdx (96), guides/solution-engineering/getting-started.mdx (196), guides/solution-engineering/rfp-responses.mdx (96)

[jordanc-relevanceai]

Superseded by #806, which lands the same idea under the name Operations and keeps Tasks and Sharing where they were.