docs(85bacd35-7888-4e9f-ba59-c586df9d5538): add trigger inventory page

[claude]

Summary

• Adds new documentation page at build/agents/build-your-agent/trigger-inventory.mdx for the trigger inventory feature
• Documents the centralized trigger view, bulk alert configuration, trigger selection with persistence across pagination, progress indicators, and error handling
• Updates docs.json to include the new page after alerts in the Build an Agent section

Linear: https://linear.app/relevance/issue/85bacd35-7888-4e9f-ba59-c586df9d5538/

Test plan

• Verify page renders correctly in the docs site
• Confirm navigation link works from the sidebar
• Check all internal links resolve (alerts page reference)
• Verify early access note is prominent and accurate

🤖 Generated with Claude Code

[mintlify]

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

Project	Status	Preview	Updated
relevanceai	🟢 Ready	View Preview	Oct 14, 2026, 11:43 PM

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

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated
relevance-docs	Ready	Preview	Oct 14, 2026 11:44pm UTC

[linear]

TSP-1468

[github-actions]

🎯 Vibe check

Reviewed: 2 files (1 with issues, 1 clean)

Scores

	Dimension	Score	What's holding it back
🔴	Consistency	5/10	"trigger/triggers" appears ~15 times as lowercase when referring to the Relevance AI product feature — should be "Trigger/Triggers" per CLAUDE.md. Same issue with "agent" on line 15.
🟢	Technical clarity	9/10	One terminology slip: the intro and heading use "alerts" but UI buttons say "alarms" — worth verifying these are the actual product labels.
🟢	Non-technical clarity	9/10	Clean prose, logical flow, no jargon dropped cold.
🟡	Structure	7/10	Two sequential procedure sections use numbered markdown lists instead of <Steps>.

Score key: 🟢 9–10, 🟡 6–8, 🔴 1–5.

✨ Overall vibe: The content is solid — clear, focused, and well-scoped for an early-access feature page. The main problem is mechanical: "trigger" is used in lowercase throughout when it refers to the product feature, which violates CLAUDE.md's capitalization rules and is easy to fix.

🔧 Issues (17)

• trigger-inventory.mdx:11 — "every trigger across all agents" → "every Trigger across all Agents" (both are product features)
• trigger-inventory.mdx:11 — "check or update trigger alerts" → "Trigger alerts"
• trigger-inventory.mdx:13 — heading "## Accessing the trigger inventory" → "## Accessing the Trigger inventory"
• trigger-inventory.mdx:15 — "The trigger inventory page" → "The Trigger inventory page"
• trigger-inventory.mdx:15 — "all triggers across your workspace" → "all Triggers across your workspace"
• trigger-inventory.mdx:15 — "regardless of which agent owns them" → "Agent"
• trigger-inventory.mdx:19 — "all workspace triggers" → "all workspace Triggers"
• trigger-inventory.mdx:23 — "alerts on multiple triggers" → "Triggers"
• trigger-inventory.mdx:25 — heading "### Selecting triggers" → "### Selecting Triggers"
• trigger-inventory.mdx:27 — "select those triggers" → "Triggers"
• trigger-inventory.mdx:28 — "select all triggers on the current page" → "Triggers"
• trigger-inventory.mdx:33 — "Select one or more triggers" → "Triggers"
• trigger-inventory.mdx:38 — "how many triggers have been processed" → "Triggers"
• trigger-inventory.mdx:44 — "Select one or more triggers" → "Triggers"
• trigger-inventory.mdx:47 — "affects triggers that already have alerts active" → "Triggers"
• trigger-inventory.mdx:56 — "the specific trigger" → "Trigger"
• trigger-inventory.mdx:8 — The <Info> callout names two feature flags but doesn't tell the user how to get them enabled. Add a link to /get-started/support so readers know where to ask (e.g., "Contact support to enable these for your workspace.").

🧩 Component suggestions (2)

• trigger-inventory.mdx:33–36 — "Setting up alerts in bulk" uses a numbered markdown list for a sequential procedure. Replace with <Steps> to add visual progress indicators. Same fix applies to "Muting alerts in bulk" (lines 44–46).

• trigger-inventory.mdx:34,38,45 — The page uses "alerts" in all prose and headings ("Bulk alert configuration", "Set up alerts", "Muting alerts") but the UI button labels are Set up alarm and Mute alarms, and the progress text reads "Enabling alarms…". If both terms are genuinely in use in the product UI, add a one-liner clarifying the relationship ("The platform uses alarms for these bulk controls — they map to the alert rules configured per trigger."). If "alarms" is wrong, align the button names to match the rest of the page.

✅ Clean files (1)

docs.json — trigger-inventory is correctly placed in the navigation at build/agents/build-your-agent/trigger-inventory, immediately after triggers. No structural issues.

🔋 Credit usage

Item	Count
Files reviewed	2
Context pages read	2
Total lines processed	~394

Files read: trigger-inventory.mdx (57 lines), docs.json (100 lines, first 100 of full file), alerts.mdx (111 lines), triggers.mdx (126 lines)

[github-actions]

🎯 Vibe check

Reviewed: 2 files (1 with issues, 1 clean)

Scores

	Dimension	Score	What's holding it back
🔴	Consistency	5/10	"Trigger" (a Relevance AI product feature) is written as lowercase "trigger" in 15+ instances across the page.
🟢	Technical clarity	9/10	Good specificity throughout — exact button labels, progress indicator behavior, and error-handling semantics are all called out. Minor: "alarms" appears in button labels while prose consistently uses "alerts", which may confuse readers cross-referencing the Alerts page.
🟢	Non-technical clarity	9/10	Clear intro, upfront early-access callout, and helpful behavioral details (selection persistence across pagination, skipping triggers-without-alerts on mute).
🟡	Structure	7/10	Sequential how-to steps in "Setting up alerts in bulk" and "Muting alerts in bulk" are plain numbered lists — the <Steps> component would add visual progress indicators that fit these flows.

Score key: 🟢 9–10, 🟡 6–8, 🔴 1–5.

✨ Overall vibe: The page is well-written and technically precise — it covers selection persistence, progress indicators, and error-handling edge cases that most feature docs skip. The one thing to fix before shipping is the capitalization of "Trigger" throughout: every reference to the product feature is lowercase, and it's everywhere.

🔧 Issues (18)

"Trigger" capitalization — product feature must be capitalized (CLAUDE.md)

• trigger-inventory.mdx:13 — heading "## Accessing the trigger inventory" → "## Accessing the Trigger inventory"
• trigger-inventory.mdx:25 — heading "### Selecting triggers" → "### Selecting Triggers"
• trigger-inventory.mdx:11 — "every trigger across all agents" → "every Trigger across all agents"
• trigger-inventory.mdx:15 — "The trigger inventory page lists all triggers" → "The Trigger inventory page lists all Triggers"
• trigger-inventory.mdx:19 — "all workspace triggers" → "all workspace Triggers"
• trigger-inventory.mdx:23 — "alerts on multiple triggers at the same time" → "alerts on multiple Triggers"
• trigger-inventory.mdx:27 — "select those triggers" → "select those Triggers"
• trigger-inventory.mdx:28 — "select all triggers on the current page" → "select all Triggers on the current page"
• trigger-inventory.mdx:33 — "Select one or more triggers using the checkboxes" → "Select one or more Triggers"
• trigger-inventory.mdx:38 — "how many triggers are selected" and "how many triggers have been processed" → "Triggers" (both)
• trigger-inventory.mdx:44 — "Select one or more triggers that already have alerts" → "Triggers"
• trigger-inventory.mdx:48 — "only affects triggers that already have alerts active", "triggers without alerts are skipped", "all selected triggers" → "Triggers" (all three)
• trigger-inventory.mdx:52 — "selecting triggers" and "next to that trigger" → "Triggers" / "Trigger"
• trigger-inventory.mdx:56 — "fails on some triggers" → "Triggers"

"Alarms" vs "alerts" terminology

• trigger-inventory.mdx:34 — button label Set up alarm and trigger-inventory.mdx:45 Mute alarms use "alarm/alarms" while the surrounding prose and the sibling Alerts page use "alerts" throughout. If these are the real button labels in the product UI, they're correct — but worth a note to the product team since the mismatch will confuse readers who just read the Alerts page.

🧩 Component suggestions (2)

• trigger-inventory.mdx:33–36 — "Setting up alerts in bulk" uses a plain numbered list for a 4-step sequential procedure. Replace with <Steps> for visual progress indicators, which fit a how-to flow better than a bare list.
• trigger-inventory.mdx:44–46 — "Muting alerts in bulk" is the same pattern — 3 sequential steps. Wrap in <Steps>.

✅ Clean files (1)

docs.json — trigger-inventory is correctly inserted after triggers in the build/agents/build-your-agent group. No structural issues.

🔋 Credit usage

Item	Count
Files reviewed	2
Context pages read	2
Total lines processed	~394

Files read: trigger-inventory.mdx (57 lines), triggers.mdx (126 lines), alerts.mdx (111 lines), docs.json (100 lines, partial)

[github-actions]

🎯 Vibe check

Reviewed: 2 files (1 with issues, 1 clean)

Scores

	Dimension	Score	What's holding it back
🟡	Consistency	6/10	Widespread lowercase "trigger", "agent", "workforce" throughout the MDX file — these are Relevance AI product terms (Trigger, Agent, Workforce) and must be capitalized. The same page correctly capitalizes "Triggers" in some sentences (lines 66, 81, 82, 84) but not in others, making it inconsistent even with itself. One British phrase ("straight away").
🟡	Technical clarity	8/10	"classic sidebar" (line 14) is unexplained — readers won't know what distinguishes it from the current sidebar, or whether they're even using it.
🟢	Non-technical clarity	9/10	No issues found.
🟡	Structure	8/10	Two sequential numbered procedures ("Setting up alarms in bulk", "Muting alarms in bulk") should use <Steps> for visual progress indicators.

Score key: 🟢 9–10, 🟡 6–8, 🔴 1–5.

✨ Overall vibe: Solid how-to page — clear prose, well-organized sections, precise UI references, and no banned words or structural red flags. The main issue is a systematic capitalization problem: "trigger", "agent", and "workforce" appear in lowercase throughout almost every prose sentence, which is inconsistent with both CLAUDE.md and other pages in the same directory (triggers.mdx caps them consistently). Fix the capitalization sweep and this is in good shape.

🔧 Issues (18)

Product term capitalization — Trigger / Agent / Workforce must be capitalized when referring to Relevance AI features:

• trigger-inventory.mdx:4 (frontmatter description) — "triggers across all agents and workforces" → "Triggers across all Agents and Workforces"
• trigger-inventory.mdx:7 — "whether it belongs to an agent or a workforce" → "Agent or a Workforce"; "every trigger in your project" → "every Trigger"
• trigger-inventory.mdx:11 — "lists every trigger in your project" → "every Trigger"
• trigger-inventory.mdx:14 — "agents and workforces" → "Agents and Workforces"
• trigger-inventory.mdx:23 — "the agent or workforce that owns the trigger" → "Agent or Workforce … Trigger"
• trigger-inventory.mdx:30 (heading ### Grouping triggers) — "triggers" → "Triggers": ### Grouping Triggers
• trigger-inventory.mdx:38 (heading ## Setting up an alarm on one trigger) — "trigger" → "Trigger": ## Setting up an alarm on one Trigger
• trigger-inventory.mdx:40 — "Click the bell in a trigger's" → "a Trigger's"
• trigger-inventory.mdx:49 — "mute alarms on many triggers at once" → "Triggers"
• trigger-inventory.mdx:51 (heading ### Selecting triggers) — "triggers" → "Triggers": ### Selecting Triggers
• trigger-inventory.mdx:53 — "select those triggers" and "select all triggers on the current page" → "Triggers" both instances
• trigger-inventory.mdx:57 — "When you select triggers, an action bar appears" → "Triggers"
• trigger-inventory.mdx:61 — "Select one or more triggers" → "Triggers"
• trigger-inventory.mdx:72 — "Select one or more triggers" → "Triggers"
• trigger-inventory.mdx:75 — "Only triggers whose alarm is currently on" → "Only Triggers"
• trigger-inventory.mdx:77 (heading ### When some triggers fail) — "triggers" → "Triggers": ### When some Triggers fail

American English:

• trigger-inventory.mdx:44 — "clicking the bell turns it off straight away" → "straight away" is British/Australian; use "immediately" or "right away"

Clarity:

• trigger-inventory.mdx:14 — "In the classic sidebar, the entry only shows for…" — "classic sidebar" is unexplained. Is this a named UI mode? If it refers to a legacy navigation layout, name it precisely or link to where users can identify which sidebar they have.

🧩 Component suggestions (2)

• trigger-inventory.mdx:61–64 — "Setting up alarms in bulk" uses a plain numbered list for a sequential 4-step procedure. Replace with <Steps> for visual progress indicators:

  <Steps>
    <Step title="Select triggers">Check one or more rows.</Step>
    <Step title="Open the alarm modal">In the action bar, click **Set up alarm**.</Step>
    <Step title="Choose notification channels">Select email, Slack, or Microsoft Teams.</Step>
    <Step title="Save">Click Save to apply the alarm to all selected Triggers.</Step>
  </Steps>

• trigger-inventory.mdx:72–73 — "Muting alarms in bulk" is only 2 steps but is still a sequential procedure. Use <Steps> for consistency with the alarm setup section above.

✅ Clean files (1)

docs.json — trigger-inventory is correctly placed in the navigation tree immediately after build/agents/build-your-agent/triggers. No structural issues.

🔋 Credit usage

Item	Count
Files reviewed	2
Context pages read	2
Total lines processed	~422

Files read: trigger-inventory.mdx (85 lines), docs.json (100 lines, first section), alerts.mdx (111 lines), triggers.mdx (126 lines)

[github-actions]

🎯 Vibe check

Reviewed: 3 files (1 with issues, 2 clean)

Scores

	Dimension	Score	What's holding it back
🟡	Consistency	7/10	Four instances of lowercase "agent"/"workforce" where the Relevance AI product names Agent/Workforce should be capitalized (lines 7, 14, 23, 28).
🟢	Technical clarity	9/10	Good specific UI references and edge-case coverage throughout. "Classic sidebar" in the Note (line 14) is used without explanation — readers won't know what differentiates it from the non-classic sidebar.
🟢	Non-technical clarity	9/10	No issues found.
🟡	Structure	8/10	Two sequential procedures use plain markdown numbered lists instead of <Steps> (lines 60–64, 71–73).

Score key: 🟢 9–10, 🟡 6–8, 🔴 1–5.

✨ Overall vibe: A clean, well-scoped reference page with accurate content and good edge-case coverage (bulk action progress, partial failures, disabled button state). Main cleanup is capitalizing Agent/Workforce as Relevance AI product names and upgrading two numbered procedures to <Steps> components.

🔧 Issues (4)

• site/content/docs/build/agents/build-your-agent/trigger-inventory.mdx:7 — "an agent or a workforce" → "an Agent or a Workforce" (these are Relevance AI product names)
• site/content/docs/build/agents/build-your-agent/trigger-inventory.mdx:14 — "create agents and workforces" → "create Agents and Workforces"
• site/content/docs/build/agents/build-your-agent/trigger-inventory.mdx:23 — "the agent or workforce that owns the trigger. Click it to open that agent or workforce." → capitalize both Agent and Workforce in each occurrence
• site/content/docs/build/agents/build-your-agent/trigger-inventory.mdx:28 — "filter by trigger, workforce or agent name" → "filter by trigger, Workforce or Agent name"

🧩 Component suggestions (2)

• site/content/docs/build/agents/build-your-agent/trigger-inventory.mdx:60–64 — "Setting up alarms in bulk" uses a markdown numbered list for a 4-step sequential procedure. Wrap in <Steps> to add visual progress indicators: <Steps><Step title="Select triggers">...</Step><Step title="Click Set up alarm">...</Step>...</Steps>
• site/content/docs/build/agents/build-your-agent/trigger-inventory.mdx:71–73 — "Muting alarms in bulk" also uses a plain numbered list. Same fix — <Steps> with two steps.

✅ Clean files (2)

docs.json (navigation entry added correctly at the right position), site/content/docs/build/agents/meta.json (trigger-inventory placed after alerts, matches docs.json order)

🔋 Credit usage

Item	Count
Files reviewed	3
Context pages read	2
Total lines processed	~348

Files read: trigger-inventory.mdx (85 lines), meta.json (26 lines), alerts.mdx (111 lines), triggers.mdx (126 lines)