[roethke]

What changed? Why?

Expands Bridge to Base beyond the two standard-bridge UIs, so readers can find faster routes and cross-chain token options.

• Compare Bridge Types (new): standard bridge vs fast bridge vs messaging protocol, by withdrawal time and trust model. Uses current withdrawal timing: 5 days, or 1 day when both TEE and ZK proofs back the proposal.
• Fast Bridges (new, H2): Across, Relay, Stargate, with a warning that fast bridges do not inherit Base's security.
• Bridge Aggregators (new, H2): Jumper, Bungee.
• Cross-Chain Messaging Protocols (new, H2): Chainlink CCIP (CCT) and LayerZero V2 (OFT) for multichain tokens, plus the LayerZero endpoint ID for Base Mainnet (30184). For Token Issuers links to it.
• Bridge Aggregators: notes that route security depends on the bridge the aggregator selects.
• Finalization window: defined and linked to the Withdrawals spec where first used.
• Choose a Route: Ethereum row points to fast bridges; new "Another EVM chain" row.
• Withdrawals spec: the faster-withdrawals paragraph links to the new Fast Bridges and Compare Bridge Types sections.

Notes to reviewers

• Providers are listed neutrally under the existing Disclaimer; no ranking.
• Base Mainnet support was checked against each provider's own source on 2026-10-19:
  • Across: available-routes API returns Ethereum → Base (8453) routes.
  • Relay: chains API lists Base (8453), enabled.
  • Stargate: LayerZero transfer API lists Base (8453).
  • Chainlink CCIP: Base mainnet network page.
  • LayerZero: deployed contracts list endpoint ID 30184 as Base Mainnet.
• Superbridge aggregates the OP Stack native bridge and third-party bridges (eg, Across, Relay), so the page no longer describes it as standard-bridge only.
• Withdrawal timing matches the Withdrawals spec and the Beryl upgrade page (live on Base Mainnet since 2026-06-25): 5 days single-proof, 1 day dual-proof.
• Fast bridge timing ("typically seconds to minutes") is a range across providers, not a guarantee.

How has it been tested?

• node scripts/lint-mdx.js on both changed pages: 0 errors, 0 warnings
• node scripts/validate-docs-structure.js and node scripts/check-terminology.js: pass
• node scripts/llms.js regenerated llms.txt / llms-full.txt
• Rendered both pages locally with mint dev at 1280px; tables fit without clipping

Screenshots

Pending upload.

[mintlify]

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

Project	Status	Preview	Updated
base	🟢 Ready	View Preview	Oct 19, 2026, 9:49 PM

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

[cb-heimdall]

✅ Heimdall Review Status

Requirement	Status	More Info
Reviews	✅ 1/1	Denominator calculation Show calculation 1 if user is bot 0 1 if user is external 0 2 if repo is sensitive 0 From .codeflow.yml 1 Additional review requirements Show calculation Max 0 0 From CODEOWNERS 0 Global minimum 0 Max 1 1 1 if commit is unverified 0 Sum 1

[github-actions]

Docs review

Risk: Medium (three AI models review independently) · risk: medium · shadow mode, not enforced

• adds or changes a chain ID or fee number in docs/base-chain/network-information/ecosystem-bridges.mdx
• adds or changes a chain ID or fee number in docs/llms-full.txt
• changes 137 lines across 4 files (low allows up to 20 lines and 3 files)

Verdict: Fits, with changes · reviewed 25e5999 with claude-sonnet-5-5

The PR expands the live Bridge to Base page with fast bridges, aggregators, and messaging protocols, and updates the Withdrawals spec link to match. The change belongs on these pages, and the added internal anchors resolve. The main fixes are a spec page that now points readers to a third-party provider list, a few unsupported or shaky claims about providers and message-protocol security, and some wording that doesn't match the neighboring pages.

Suggestions

1. Spec page now links into third-party provider list (placement, docs/specifications/base-protocol/bridging/withdrawals.mdx L23) — suggestion posted inline
   Spec pages are normative and should avoid pointing readers into provider marketing-style lists. The sentence is acceptable, but the link text 'Fast Bridges' reads as a Base-endorsed feature. Link to the comparison section and keep the provider list as a secondary pointer.
   Evidence: Content guidelines: spec pages are 'normative, not tutorial' and 'Link, don't duplicate'. The Disclaimer on the bridge page says providers are independent third parties.
2. 'Pay you out' liquidity description overgeneralizes (accuracy, docs/base-chain/network-information/ecosystem-bridges.mdx L67) — suggestion posted inline
   The sentence says all fast bridges pay out from destination liquidity and then settle on their own. Stargate uses pooled liquidity and Relay uses solvers, so the generalization may not hold for every provider. It also uses a second-person 'pay you out' framing. Describe the shared pattern more neutrally.
   Evidence: The table directly below describes different mechanisms per provider (intent-based relayers, solvers, liquidity pools).
3. Security statement for messaging protocols is too absolute (accuracy, docs/base-chain/network-information/ecosystem-bridges.mdx L103) — suggestion posted inline
   The claim that a token is secured by 'the protocol's own network' is stated as fact for both protocols. Security differs between CCIP and LayerZero (for example, LayerZero depends on configurable verifier settings). Say the token's security depends on the protocol and its configuration, which also matches the table row above.
   Evidence: The table at line 32 says trust is 'The protocol and the token's settings', so the closing sentence is inconsistent with it.
4. Standard bridge and Superbridge guidance is split across sections (consistency, docs/base-chain/network-information/ecosystem-bridges.mdx L49) — suggestion posted inline
   The new paragraph after the provider cards mixes the withdrawal timing with a Superbridge routing caveat. Compare Bridge Types already covers the timing. Keep the paragraph short and split the two ideas into separate sentences with active voice.
   Evidence: Writing rules: one idea per sentence, active voice, concise. Compare Bridge Types already states the finalization window.
5. Intro sentence is weak and 'trust' is unexplained (writing, docs/base-chain/network-information/ecosystem-bridges.mdx L26) — suggestion posted inline
   'fall into three types. They differ in speed and in who you trust' is vague, and the table says 'You trust' without defining the trust assumption. Lead with the concrete difference.
   Evidence: Tone guidelines: strong verbs, data over description, define terms on first use.
6. Redundant link and wordy closing sentence (consistency, docs/base-chain/network-information/ecosystem-bridges.mdx L34) — suggestion posted inline
   The Withdrawals page is linked twice within a few lines, and 'To see why the window exists, read Withdrawals' is a tutorial-style pointer. Fold the scope statement into one sentence and keep one link.
   Evidence: Table at line 30 already links the finalization window to the same Withdrawals page.
7. 'Another EVM chain' row recommends unvetted routes without caveat (accuracy, docs/base-chain/network-information/ecosystem-bridges.mdx L14)
   The Choose a Route table lists fast bridges and aggregators as the recommended route for any EVM chain. Support for Base was only verified for specific providers, and the aggregators were not listed in the PR's verification notes. Either confirm Jumper and Bungee support Base Mainnet or soften the claim.
   Evidence: PR notes list verification for Across, Relay, Stargate, CCIP and LayerZero, but not Jumper or Bungee.
8. Intro line claims support without a source for CCIP wording (writing, docs/base-chain/network-information/ecosystem-bridges.mdx L94) — suggestion posted inline
   'Both of these support Base Mainnet' is a bare claim; the table links already show it. Use a direct statement of what the section covers instead.
   Evidence: Tone guidelines: no preamble, state facts directly.

What the reviewer checked

Page	Placement	Index section	Neighbors read
docs/base-chain/network-information/ecosystem-bridges.mdx	Live in navigation	not in llms-full.txt	0
docs/specifications/base-protocol/bridging/withdrawals.mdx	Live in navigation	not in llms-full.txt	0

Links added: /specifications/base-protocol/bridging/withdrawals (live), /base-chain/network-information/ecosystem-bridges#fast-bridges (live), /base-chain/network-information/ecosystem-bridges#compare-bridge-types (live)

Advisory review from the Docs PR review workflow (run). It does not approve, block, or edit this PR. Add the skip-ai-review label to turn it off for this PR.

[github-actions]

Suggested edits from the docs review. Each one is optional; apply it with "Commit suggestion" if you agree.

[github-actions]

Suggested edits from the docs review. Each one is optional; apply it with "Commit suggestion" if you agree.

[github-actions]

Suggested edits from the docs review. Each one is optional; apply it with "Commit suggestion" if you agree.

Approved review 5578079507 from soheimam is now dismissed due to new commit. Re-request for approval.

[github-actions]

Suggested edits from the docs review. Each one is optional; apply it with "Commit suggestion" if you agree.