quest: plan epoch correctness fixes and IETF extension

[kixelated]

Problem

The epoch audit reproduced JS requests losing a covering prefix's epoch and reusing a cached broadcast after a more-specific announcement wins. Existing quests also need a consistent plan for legacy clients and negotiated IETF identity.

Approach

• Add an m0 JS consume-identity quest and move the existing sticky-request parity fix into the broadcast-epoch release gate.
• Keep the catalog restart and follower-gap fixes with their existing owners (fix(net)!: the application owns a replacement track's start sequence #5068 and fix(moq-net): a followed path reports a gap onto a same-epoch prefix #5188). Update the TS export quest to use the shared path follower; feat!: TS export lingers within an epoch, --stitch switches programs #5147 still has a separate prefix watcher.
• Add an m1 IETF epoch extension for drafts 17-22, retaining paths and epochless clients on updated relays.
• Extend claim-epoch planning with a separate IETF follow-up. The existing quest owns the shared model and lite-07 wire, so the lite-07 cut does not depend on IETF.

Validation: nix develop --command just check, quest check (439 documents), and git diff --cached --check passed. This PR changes plans only; the audit reproductions inform the regression scenarios in the quests.

Impact

• Public API: none in this PR. Planned work corrects JS request/cache behavior and reuses the shared extension configuration and epoch model.
• Wire: none in this PR. Planned IETF changes require negotiation; older peers retain their framing and paths. Updated relays retain identity internally and translate replacements into End+Start for epochless clients.
• Compatibility: End+Start is not general cache invalidation under IETF's immutable object names. Arbitrary older caching relays and clients retaining objects across broadcasts are outside the guarantee.

Alternatives

Rejected epoch path suffixes, mandatory epoch support for clients, draft-22-only support, making the extension a release blocker, and requiring a separate migration guide. Seamless JS handover stays in m1.

Follow-ups

Implementation remains in the linked quests. The extension implementation owns its normative draft and updates to the cluster draft and existing identity/compatibility docs. No unrelated repository changes are included.

Decision paper trail

1. Goal: plan the audit fixes and a path-preserving negotiated IETF extension as separate work, reconciling existing quests. ✅ Full scope / Narrower audit-only scope. The maintainer explicitly requested the extension.
2. Priority: ✅ Correctness fixes in m0; IETF extension and seamless JS handover in m1 / All three as m0 release blockers.
3. IETF versions: ✅ Drafts 17-22 using existing negotiation infrastructure / Draft 22 only initially.
4. Peers without the extension: ✅ Preserve legacy connectivity with epochless behavior at the boundary / Refuse epoch-bearing broadcasts across unsupported connections. The maintainer required epochless client support and then clarified the boundary: ✅ Older clients on updated relays / Also arbitrary older caching relays requiring immutable names and discovery or translation.
5. Claim-served identity: ✅ Extend the existing claim-epochs work to lite-07 and IETF as separate follow-up work / Keep claim epochs lite-only for now. Split protocol ownership to preserve the independent lite-07 cut.
6. Documentation: ✅ Include the normative extension draft and update existing identity/compatibility docs within the feature work / Also create a separate migration-guide quest.
7. Scope narrowed during the interview: ✅ Public MoQ work only; omit the initially requested downstream-repository follow-through. No downstream edits or planning PR.

The maintainer confirmed the full public scope, selected epochless clients on updated relays, and answered “all recommended” for the remaining priority, version, claim, and documentation choices.

(Written by GPT-6)

[chatgpt-codex-connector]

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-10-10T13:33:30.496183Z	ab9c5f1	Draft marked ready

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

• Configuration used: Organization UI
• Review profile: CHILL
• Plan: Advanced
• Run ID: afc1d3eb-c445-4206-8d82-9e13e4b0f35b

---

📥 Commits

Reviewing files that changed from the base of the PR and between f7df075 and ab9c5f1.

---

📒 Files selected for processing (8)

• quest/m0/broadcast-epoch/README.md
• quest/m0/broadcast-epoch/export-ts.md
• quest/m0/broadcast-epoch/js-consume-identity.md
• quest/m0/broadcast-epoch/js-restart-keeps-request.md
• quest/m1/README.md
• quest/m1/claim-epochs.md
• quest/m1/ietf-claim-epochs.md
• quest/m1/ietf-epochs.md

---

Included review availability: This review used your included allowance. Your plan provides up to 4 included reviews per hour; 1 remain after this review.

---

---

Walkthrough

The changes update m0 plans for broadcast identity with epochless clients, JavaScript request identity, and route-following behavior. They define m1 work for shared claim epochs and separate IETF broadcast-epoch and claim-epoch extensions, including compatibility boundaries and planned verification.

Estimated code review effort

Priority: ⬇️ Low

Merge Risk: ⚪ Minimal · up to ab9c5

This change records planned work; no current product behavior change or actionable merge-blocking risk is identified.

Pre-merge checks | 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Title check		The title clearly summarizes the main changes: planning epoch-correctness fixes and the IETF extension.
Description check		The description directly explains the reproduced problems, planned fixes, IETF scope, compatibility behavior, validation, and follow-up work.
Docstring Coverage		No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check. Docstring coverage is scoped to functions touched by this diff. Analyzed 0 functions across 0…
Linked Issues check		Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check		Check skipped because no linked issues were found for this pull request.

---

✨ Finishing Touches

✨ Simplify code

• Commit to this branch
• Create a new PR

---

• Autofix · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

---

_{Comment @coderabbitai help to get the list of available commands.}

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: ab9c5f18f9

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[kixelated]

Merge summary: the confirmed plan puts identity correctness and JS restart parity in m0, the negotiated IETF epoch extension for drafts 17-22 in m1, and IETF claim-response identity in a separate follow-up. Epochless clients on updated relays retain their paths and receive End+Start; arbitrary cache-retaining peers remain outside the compatibility guarantee. No production API or wire changes are made by this PR.

Local just check and quest validation passed. CodeRabbit reported no actionable findings on ab9c5f1. Codex reviewed that same head; its legacy-cache concern has been answered with the maintainer's confirmed scope and the limitation already documented in the plan. There are no remaining decisions. Enabling auto-merge for that exact head, gated by required CI and the merge queue.

(Written by GPT-6)