docs(drift): draft from internal source

[pedromcunha]

Drafted from an internal source repo, requested by @cucupac.

This PR was auto-generated by the reusable-docs-draft-pr workflow from a Claude-drafted update in the source repo. It is opened as a draft — please review, edit as needed, and mark ready for review before merging.

Review checklist

• Content matches the behavior shipped in the source PR
• No internal-only details (service names, file paths, unreleased names, security internals) leaked into the draft
• Conforms to CLAUDE.md style rules
• Cross-references and internal links resolve
• Any Needs human input items from the source comment are addressed

---

Generated by reusable-docs-draft-pr.yml in relayprotocol/agents-config.

[mintlify]

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

Project	Status	Preview	Updated (UTC)
unevenlabs	🟢 Ready	View Preview	Jun 16, 2026, 3:21 AM

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

[github-actions]

Claude finished @pedromcunha's task in 1m 44s —— View job

---

Code Review — PR #340

• Read the diff and surrounding context
• Check conformance to CLAUDE.md style rules
• Verify cross-references / internal links resolve
• Check for leaked internal-only details
• Post review feedback

---

This is a clean, well-placed addition to the Handling Quote Regeneration section of features/deposit-addresses.mdx. It documents data.supersededByRequestId as a direct pointer from an original request to the regenerated request that owns the fill lifecycle, and the <Tip> nicely contrasts it with the existing includeChildRequests=true approach. Overall it conforms to the style guide. A few notes:

✅ Style conformance (CLAUDE.md)

• Paragraph length (§2.10): the new prose is 2 sentences — within the 2–3 limit. 👍
• Callout length (§2.5): the <Tip> is 2 lines, well under the 15-line cap and contains no nested code/tables. 👍
• Emphasis (§2.9): code values (requestId, data.supersededByRequestId, includeChildRequests=true) are correctly backticked in prose.
• Placeholders (§2.8): <ORIGINAL_REQUEST_ID> / <REGENERATED_REQUEST_ID> use the all-caps angle-bracket style already established on this page (<DEPOSIT_ADDRESS>, <REQUEST_ID>), so they're consistent.
• JSON shape: the example's top-level id / status / data structure matches the documented request object shape elsewhere on the page (around features/deposit-addresses.mdx:963), so it reads as a faithful trimmed response.

🔍 Please verify before marking ready

1. Field accuracy (review-checklist item 1). Since this was auto-drafted from an internal source, confirm the field is literally returned as data.supersededByRequestId on the requests/v2 response (exact casing and nesting under data), and that the original request's status genuinely stays pending/its original public status after the handoff. This is the one claim a reviewer can't validate from the docs repo alone.
2. Internal-detail phrasing (checklist item 2). The line "the actual sweep and fill move under the regenerated request" describes internal lifecycle mechanics. It reads as integrator-facing API behavior (not a leaked service name/path), but please confirm "sweep" and "fill lifecycle" are terms already used publicly in these docs — a quick scan shows "sweep" is used elsewhere on this page, so this is likely fine.

💡 Optional nit (consistency, non-blocking)

The new bash + json pair is presented as two bare fenced blocks. The very next section (Request-Level Status, features/deposit-addresses.mdx:1265) pairs its request/response in a <CodeGroup> (bash Request / json Response). For local consistency you could wrap the new pair the same way. Note the rest of the Handling Quote Regeneration section also uses a bare bash block, so this isn't a regression either way — purely a polish call.

No broken links or frontmatter issues. No changes were made — this is review-only per the request.
• docs-drift/src-1db97dd688-4c225b5

[github-actions]

Claude finished @pedromcunha's task in 57s —— View job

---

Code Review — PR #340

• Read the diff and surrounding context
• Check conformance to CLAUDE.md style rules
• Verify cross-references / internal links resolve
• Check for leaked internal-only details
• Post review feedback

---

The diff adds a 24-line block to the Handling Quote Regeneration section of features/deposit-addresses.mdx, documenting data.supersededByRequestId as a direct pointer from an original request to the regenerated request that owns the fill lifecycle, plus a <Tip> contrasting it with the existing includeChildRequests=true approach. (This sync trigger came from a merge of main into the branch — the documented content is unchanged from the prior review.) It conforms to the style guide. Details below.

✅ Style conformance (CLAUDE.md)

• Paragraph length (§2.10): the new prose at features/deposit-addresses.mdx:1237 is 2 sentences — within the 2–3 limit.
• Callout length (§2.5): the <Tip> is 2 lines, well under the 15-line cap, with no nested code/tables. <Tip> is the correct component for a "when to use which" pointer (§2.5).
• Emphasis (§2.9): code values (requestId, data.supersededByRequestId, includeChildRequests=true, supersededByRequestId) are correctly backticked in prose.
• Code examples (§2.8): placeholders <ORIGINAL_REQUEST_ID> / <REGENERATED_REQUEST_ID> use the angle-bracket style already established on this page (<DEPOSIT_ADDRESS>, <REQUEST_ID>), so they're locally consistent even though §2.8's canonical examples are bare snake_case.
• JSON shape: the trimmed id / status / data response matches the documented request-object shape elsewhere on the page (the full example around features/deposit-addresses.mdx:962), so it reads as a faithful subset.
• Links/frontmatter: no new internal links introduced, nothing to resolve; no frontmatter touched. No broken references.

🔍 Please verify before marking ready (can't be validated from the docs repo)

1. Field accuracy (checklist item 1). Confirm the field is literally returned as data.supersededByRequestId on the requests/v2 response — exact casing and nesting under data — and that the original request's status genuinely remains pending (or its original public status) after the handoff. This is the one claim a docs-repo reviewer cannot check.
2. Internal-detail phrasing (checklist item 2). The line "the actual sweep and fill move under the regenerated request" describes lifecycle mechanics. It reads as integrator-facing API behavior rather than a leaked service name/path, and "sweep" is already used publicly elsewhere on this page (e.g. features/deposit-addresses.mdx:1285), so it's likely fine — please confirm "fill lifecycle" is acceptable public phrasing.

💡 Optional nit (non-blocking)

The new bash + json pair (:1239–:1255) is presented as two bare fenced blocks. The very next section, Request-Level Status (:1265), pairs its request/response in a <CodeGroup> with bash Request / json Response labels. Wrapping the new pair the same way would improve local consistency. That said, the rest of the Handling Quote Regeneration section uses a bare bash block, so leaving it as-is isn't a regression — purely polish. Fix this →

No changes made — this is review-only.
• docs-drift/src-1db97dd688-4c225b5