Skip to content

docs(tip-1094): rewrite for readability - #9502

Open
0xrusowsky wants to merge 3 commits into
tip/1094from
rusowsky/tip-1094-readability
Open

0xrusowsky wants to merge 3 commits into
tip/1094from
rusowsky/tip-1094-readability

Conversation

@0xrusowsky

Copy link
Copy Markdown
Contributor

Restructures TIP-1094 to follow the deposit flow: address format, key versions, registration, issuance, sending, zone processing, refunds and results. Drops the Alternatives considered section, which already lists its rejected options inline where they matter.

Normative keyword counts are unchanged (MUST 25, MUST NOT 19, SHOULD 4, SHOULD NOT 1, MAY 1). Interfaces and Invariants are unchanged, except that the portal minimum-version storage slot moved from the TIP-1180 bullet to Storage layouts.

@tempoxyz-bot

tempoxyz-bot commented Nov 21, 2026 •

Copy link
Copy Markdown
Contributor

👁️ Cyclops Security Review

✅ Reviewed 2867395 (latest) · Coverage: partial · no open findings · report
⚠️ Not checked: The replacement relative path zones/specs/spec.md is absent at the pinned Tempo head, and the previously linked tempoxyz/zones specification was not staged. The inherited ECIES, queue, and refund requirements could not be checked against th
⚠️ Not checked: TIP-1094 transfer routing, portal admission and version retirement specification: ruled out without reading the cited code at this commit
⚠️ Not checked: TIP-1094 queue authentication, rejection and settlement specification: ruled out without reading the cited code at this commit
⚠️ 1 more unchecked question is listed in the report
ℹ️ Not verifiable from the source alone (needs running code or deployed state): The draft does not establish deployed T15 activation, implementation conformance, or release-build worst-case resource costs for TIP-1094. To verify it: Provide pinned TIP-1094 validator, prover, port

@claude claude Bot left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Nothing blocking. The comments below are optional suggestions. There is no need to push a fix for them before merging.

Comment thread tips/tip-1094.md
Comment on lines 20 to 26
## Motivation

Zones accept user deposits only through the portal's `deposit()` and `depositEncrypted()` calls ([Zones protocol](https://github.com/tempoxyz/zones/blob/a88c2ac6b6b407225e2861c660806faccc0217df/specs/spec.md#deposits)). Both require the sender to build an ECIES payload that encrypts the recipient and memo to the sequencer's key. Many systems (wallets, exchanges, and payment systems) can send ordinary TIP-20 transfers, but would need integration work to build that payload and call the portal.
Zones accept user deposits only through the portal's `deposit()` and `depositEncrypted()` calls ([Zones protocol](../zones/specs/spec.md#deposits)). Both require the sender to build an ECIES payload that encrypts the recipient and memo to the sequencer's key. Wallets, exchanges and payment systems can send ordinary TIP-20 transfers, but would need integration work to build that payload and call the portal.

This TIP gives each zone recipient the option to have reusable addresses on mainnet that work with a plain transfer and do not reveal who the recipient is. The central idea is to encrypt the recipient's identifier into the address itself. The zone registers the recipient privately and assigns it a 4-byte master ID. The recipient adds a 2-byte user tag to tell senders apart. The zone encrypts those 6 bytes with FF1, a standard NIST cipher whose output is the same size as its input to generate a ciphertext. Only the zone holds the key, so only the zone can recover the master ID and credit the recipient.
This TIP gives each zone recipient reusable addresses on Tempo that work with a plain transfer and do not reveal the recipient. The zone registers the recipient privately and assigns it a 4-byte master ID. The recipient adds a 2-byte user tag to tell senders apart. The zone encrypts those 6 bytes with FF1, a NIST cipher whose output is the same size as its input, and puts the ciphertext in the address. Only the zone holds the key, so only the zone can recover the master ID and credit the recipient.

Because addresses are computed from the master ID and tag, there is no per-address state. Registration and lookup stay in confidential zone state, so Tempo mainnet never learns that a recipient registered. Each zone has its own registry; receiving in several zones takes one registration in each.

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

🟡 (optional) The rewritten TIP no longer documents alternatives considered, which the tip-0000 draft gate requires. The reason for not reusing the TIP-1022 registry is also gone. The deleted subsection was the only place that said reusing the L1 TIP-1022 registry would make registrations public and link master IDs to recipients. It also covered keeping only ECIES deposits and exposing the master ID, and nothing else in the TIP covers those. Fix: put the alternatives back, either as a short subsection or inline in Motivation as the template suggests. Each rejected option needs its reason, especially the TIP-1022 registry privacy argument.

Why this was flagged

AGENTS.md says a TIP must follow tips/tip-0000.md's process and quality gates. tips/tip-0000.md:59 requires the draft to include "alternatives considered", and tips/tip_template.md's Motivation asks for alternatives where they apply. The diff deletes the "### Alternatives considered" subsection at the base's Motivation. The PR says these options are listed inline already, but only the per-token registration option is (line ~90). The remaining TIP-1022 mentions (lines 123, 202, 382) say the registries are separate, not why. Nothing states the privacy reason: the registration is public on Tempo, so anyone could link a deposit address's master ID to its recipient. Rejecting "keep only ECIES" and "expose the master ID" is not stated anywhere either. Reviewers at the tip-0000 review and approval stages lose the record of why the design was chosen, which they have on the base branch.

Verification: nit. The diff deletes the ### Alternatives considered subsection under Motivation. tips/tip-0000.md:59 lists "alternatives considered" as part of a complete draft. The remaining TIP-1022 mentions (lines 123, 202, 382, 492) cover magic bytes, the separate registry address and rejection rules. None says why reusing the L1 registry was rejected.

This branch has not been deployed

No deployments
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants