Add AI-Assisted Debugging with Inventor documentation for Evals and Monitors

[claude]

Summary

• Adds a new AI-assisted debugging with Inventor section to evals.mdx, covering the Fix button workflow for individual Check failures and the Review with Inventor button for holistic analysis on test detail pages and Monitor dashboards
• Updates the Viewing conversation details subsection in evals.mdx to surface the Fix button
• Updates the Monitor section in evals.mdx to mention Review with Inventor on dashboards
• Adds a Use Inventor to iterate faster card to the existing Best Practices section in evals.mdx
• Adds an <Info> callout to create-an-agent.mdx cross-linking to the new Evals debugging section

Linear issue

https://linear.app/relevance/issue/TSP-1307/

Related PR

RelevanceAI/platform#15612

Test plan

• Verify all headings are in sentence case
• Verify internal anchor link /build/agents/build-your-agent/evals#ai-assisted-debugging-with-inventor resolves correctly
• Confirm feature flag callout (<Info>) renders without attributes
• Check that the new Best Practices card renders in the <CardGroup>
• Confirm no banned words (powerful, seamlessly, effortlessly, unlock, supercharge, robust, leverage)

🤖 Generated with Claude Code

[mintlify]

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

Project	Status	Preview	Updated (UTC)
relevanceai	🟢 Ready	View Preview	Jun 17, 2026, 3:47 AM

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

[linear]

TSP-1307

[github-actions]

🎯 Vibe check

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

Scores

	Dimension	Score	What's holding it back
🟡	Consistency	8/10	Bold label inside <Info> callout (evals.mdx:8); "check's" not capitalized as product term (evals.mdx:256); bold text used as section heading instead of a proper ### (evals.mdx:292).
🟢	Technical clarity	9/10	Both pages are specific with exact UI element names, detailed field tables, and gotchas (truncation behavior, sample rate guidance, credit calculation). The "check's" typo is a minor catch.
🟢	Non-technical clarity	9/10	Concepts are defined before instructions begin. Concrete example scenarios in accordions, plain-English labels throughout. No jargon dropped cold.
🟡	Structure	8/10	create-an-agent.mdx "What's next?" doesn't link to Evals — the page this PR adds. Bold pseudo-heading in the Inventor tips section breaks heading hierarchy.

Score key: 🟢 9–10, 🟡 6–8, 🔴 1–5.

✨ Overall vibe: Both pages are well-written and technically thorough — evals.mdx in particular is comprehensive, well-layered, and makes good use of tables and accordions. The main friction is a handful of style violations (bold-as-heading, bold label in callout) and one notable structural gap: the PR adds an Evals page but doesn't add it to the "What's next?" links in create-an-agent.mdx, which is the most natural place a reader would discover it.

🔧 Issues (3)

• build/agents/build-your-agent/evals.mdx:8 — **Rollout Status**: is a bold label inside an <Info> callout. CLAUDE.md says callouts must be a single short paragraph with no bold labels inside. Rewrite as a plain sentence, e.g. "Evals is rolling out progressively, starting with Enterprise customers. Reach out to your account manager to discuss access."

• build/agents/build-your-agent/evals.mdx:256 — "that check's details" — "check" should be "Check" to match the capitalized product term used everywhere else on this page.

• build/agents/build-your-agent/evals.mdx:292 — **Getting the most out of Inventor in Evals:** is bold text standing in for a section heading. CLAUDE.md says don't use bold text for section titles — use a proper heading. Change to ### Getting the most out of Inventor in Evals (no trailing colon).

🧩 Component suggestions (1)

• build/agents/build-your-agent/evals.mdx:97–99, 116–118 — Raw <div style={{...}}> wrappers around Supademo <iframe> elements. This is custom HTML with inline styles, not a Mintlify-native component. Not seen elsewhere in this repo — verify it renders as expected in the Mintlify preview before merging.

🏗️ Page structure (1)

• build/agents/create-an-agent.mdx:86–95 — "What's next?" lists Prompt, Tools, Triggers, Alerts, and Memory, but omits Evals — the page this PR adds. A reader who just created their first Agent has no path from here to testing it. Add a line: - [Evals](/build/agents/build-your-agent/evals) — test and monitor your Agent's performance with scenario-based evaluations.

⚠️ Contradictions (0)

No contradictions found between the changed files and context pages.

🔋 Credit usage

Item	Count
Files reviewed	2
Context pages read	1
Total lines processed	~762

Files read: build/agents/build-your-agent/evals.mdx (437 lines), build/agents/create-an-agent.mdx (95 lines), get-started/core-concepts/agents.mdx (230 lines)