docs(limits): integrate API platform limits via x-mint (§8, §7)

[mallikarjun-dhulange]

Platform limits — API POD (§7, §8) + deferred API-surface limits

Covers the MoEngage API rate limits (Confluence §8), partner ingestion (§7), and the API-surface limits deferred from earlier PODs. Rate limits on endpoints use the x-mint.content → #### Rate Limit convention. mint validate passes (the one warning is pre-existing and unrelated: ProfileCalculator.jsx react import).

Added

• Personalize Fetch API = 10,000 RPM → x-mint block on the fetchExperience endpoint in personalize-experience.yaml. (The 25-experience-keys-per-request cap was already documented.)

Verified already-documented (no change needed)

• api/introduction.mdx rate-limits table: Campaigns (5/25/100), Catalog (100/min·1000/hr), Campaign Stats (100/min), Custom Segments Filter (50/200/1000) and File (10/hr·100/day) — the segment API limits deferred from the Segmentation POD are already here.
• Cards Fetch & Delete API = 50K RPM (cards.yaml).
• Recommendations payload 1 MB; Catalog payload 5 MB; Track User payload 128 KB + 10K updates/min; Merge Users rate 1000/min.

⚠️ Conflicts — held for PM (see PR comment)

• User merge per operation: spec says 100 users/call, Confluence §2 says 50.
• Bulk/Track User payload: spec says 128 KB (with matching 413 error examples), Confluence §8 says 100 KB.

Both are documented in the specs with detailed surrounding content; changing the numbers would contradict the specs' own error responses, so they're flagged rather than overwritten.

Deferred — no home page exists yet

• §7 Partner ingestion (Segment 500 RPS, Adobe 500 RPS, mParticle 15K RPM, 128 KB payload): there are no CDP inbound-data pages under partner-guide/ for Segment/Adobe/mParticle. The Confluence handover itself notes these belong on "connectors pages (when created)." These limits need a home page before they can be documented.

🤖 Generated with Claude Code

[mallikarjun-dhulange]

🚩 Two spec-vs-Confluence conflicts held for PM

Both are already documented in the OpenAPI specs with values that differ from Confluence, and both have detailed surrounding content (error examples, guidance). I did not overwrite them, because changing the number would contradict the spec's own 413/error responses.

Limit	Spec (current)	Confluence	Where
User merge per operation	100 users/call	50	data.yaml /customer/merge
Bulk / Track User payload	128 KB	100 KB	data.yaml Track User (413 example says 128 KB)

Please confirm the correct values. If Confluence is right, the spec's error examples need updating too — not just the number.

Deferred — §7 partner ingestion has no home

The Segment / Adobe / mParticle inbound-ingestion limits (500 RPS, 15K RPM, 128 KB) have no CDP data-ingestion pages under partner-guide/ to live on. The Confluence doc itself flags these for "connectors pages (when created)." Once those pages exist, this is a quick follow-up.

mint validate passes for the spec change (the single warning is pre-existing: ProfileCalculator.jsx react import, unrelated to this PR).

[mintlify]

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

Project	Status	Preview	Updated (UTC)
moengage	🟢 Ready	View Preview	Jul 24, 2026, 6:33 AM

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

[mintlify]

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

Project	Status	Preview	Updated (UTC)
moengage	🟡 Building	–	Jul 24, 2026, 6:24 AM

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