id: design: postcard wire schema v1 (from #5674) #5808

Activity

1. dezren39

   added

   app/idPackage/app: id (pkgs/ or apps/ in this repo)Package/app: id (pkgs/ or apps/ in this repo)

   on Nov 5, 2026
2. developing-today commented on Nov 5, 2026

   

   developing-todayBot

   on Nov 5, 2026 – with developing.today

   Contributor

   More actions

   CODE-1491
3. dezren39

   added

   queue/queuedHas a queued task in the Supabase agent queueHas a queued task in the Supabase agent queue

   task/triageQueue packet kind: verify, reproduce, and route this itemQueue packet kind: verify, reproduce, and route this item

   on Nov 5, 2026
4. dezren39 commented on Nov 5, 2026

   

   dezren39

   on Nov 5, 2026

   MemberAuthor

   More actions

   🤖 Agent note · done

   • Agent: queue worker (repo-worker-code), lane code-i
   • Session / ID: task 10211
   • Found via: queue claim
   • Task: Triage of issue id: design: postcard wire schema v1 (from #5674) #5808 — complete. Design ticket; ordering and evidence recorded below. No implementation follow-up enqueued (the design itself is the work, and it is gated on the parent).
   • Expectation: design proceeds once parent issue id: design: Wire format & transport adapters — postcard schema, non-UDP carriers, relays/proxies #5674 records its decisions; see the scoped inputs below.
   • Links: issue id: design: postcard wire schema v1 (from #5674) #5808 · parent issue id: design: Wire format & transport adapters — postcard schema, non-UDP carriers, relays/proxies #5674 · epic issue epic: ID platform expansion — apps, comms, currency, hosting, boards (2026-10-28 directive, LIST 6–15) #2056

   Triage findings (2026-11-05, verified against current main):

   • Parent state: issue id: design: Wire format & transport adapters — postcard schema, non-UDP carriers, relays/proxies #5674 is still OPEN and parked/assigned — its surrounding decisions (wire-format family, transport-adapter model) are not yet recorded. This ticket explicitly must not re-decide them, so its decision record comes after (or with) the parent's; that is the ordering blocker, recorded here per the ticket's own terms. The ticket also states implementation timing is a separate decision, possibly months out.
   • Existing practice the design must ratify (not invent): postcard is already the live wire encoding across pkgs/id — src/protocol.rs (MetaRequest/MetaResponse over QUIC, ALPN /iroh-meta/3), src/meta_client.rs, src/peer_announcement.rs (magic prefix separating authenticated v2 gossip frames from legacy postcard payloads), and src/match_audit.rs (field order/meaning changes require a new format version). The append-only enum-variant rule (postcard uses positional discriminants) is already documented in .opencode/plans/2026-03-21_peer-discovery.md.
   • Gap the design owns: there is no dedicated wire-schema v1 record and no golden-vector corpus for the meta-protocol envelope — golden vectors exist for chain checkpoints (src/chain.rs) but not for the wire envelope/discriminators this ticket names. The decision record expected by this ticket (envelope, discriminators, golden vectors, rejected alternatives, invariants) is therefore real, unstarted work.
   • Next action: when issue id: design: Wire format & transport adapters — postcard schema, non-UDP carriers, relays/proxies #5674's decision record lands, this ticket's design is unblocked and should be worked as a design task producing the doc/ decision record; no code task should be filed from it before that record exists.
5. dezren39

   added

   status/triagedTriaged and routed (see agent note)Triaged and routed (see agent note)

   and removed

   queue/queuedHas a queued task in the Supabase agent queueHas a queued task in the Supabase agent queue

   on Nov 5, 2026
6. github-actions

   assigned

   dezren39

   on Nov 5, 2026
7. developing-today

   marked this as blocked on Nov 8, 2026

   • id: design: Wire format & transport adapters — postcard schema, non-UDP carriers, relays/proxies #5674
8. dezren39 commented on Nov 8, 2026

   

   dezren39

   on Nov 8, 2026

   MemberAuthor

   More actions

   🤖 Agent note · triaged — task 12309 (queue repo-health-code-pilot, lane code-y)

   • Verdict: real and current; still gated on the parent, exactly as the ticket's own terms require. Doctrine re-triage (2026-11-07, queue: no reconciliation between open issues and queue coverage — non-creator filings never get queue items #7016) of the task 10211 triage. id: design: postcard wire schema v1 (from #5674) #5808 asks for the postcard wire schema v1 decision record (envelope, discriminators, golden vectors) and explicitly must not re-decide the parent id: design: Wire format & transport adapters — postcard schema, non-UDP carriers, relays/proxies #5674 decisions it depends on.
   • Verified fresh (2026-11-08T05:18:25Z, this triage): issue id: design: postcard wire schema v1 (from #5674) #5808 OPEN, no held / user/owned / do-not-merge / do-not-auto-edit labels. Parent id: design: Wire format & transport adapters — postcard schema, non-UDP carriers, relays/proxies #5674 still OPEN — currently queue/claimed / work/assigned / status/in-progress with its own remainder triage (task 12244) handed off 2026-11-08T04:07Z — and its decision record has NOT landed. At laptop main b7b7f59, doc/ contains no wire-schema v1 record (postcard hits are the peer-discovery and groups/delegation docs, not this schema); the golden-vector gap for the meta-protocol envelope named by task 10211 is unchanged.
   • Why no fix packet from this triage: the design cannot be written before the parent's wire-format-family / transport-adapter decisions exist — a packet now would either stall as blocked or silently fork the parent's decisions, the exact failure the ticket forbids. The gating work is already in the queue under the parent (task 12244), so the doctrine's return-to-queue rule is carried there; id: design: postcard wire schema v1 (from #5674) #5808's own packet becomes scopable the moment id: design: Wire format & transport adapters — postcard schema, non-UDP carriers, relays/proxies #5674's decision record lands.
   • Next action (named): when id: design: Wire format & transport adapters — postcard schema, non-UDP carriers, relays/proxies #5674 records its decisions, enqueue the design packet for id: design: postcard wire schema v1 (from #5674) #5808 (decision record in doc/: chosen envelope/discriminator shape, rejected alternatives, invariants, golden-vector corpus for the envelope). This task completes as triaged; issue id: design: postcard wire schema v1 (from #5674) #5808 stays OPEN.
9. developing-today commented on Nov 8, 2026

   

   developing-todayBot

   on Nov 8, 2026 – with developing.today

   Contributor

   More actions

   🤖 Agent note · triaged — task 12309 (queue repo-health-code-pilot, lane code-z)

   • Verdict: REAL AND CURRENT, but parent-gated. Issue id: design: postcard wire schema v1 (from #5674) #5808 asks for the postcard wire schema v1 decision record (envelope, discriminators, golden vectors). That record does not exist, and the ticket's own terms forbid re-deciding the parent decisions it depends on.
   • Verified fresh (2026-11-08, this re-triage): issue id: design: postcard wire schema v1 (from #5674) #5808 OPEN; no held / user/owned / do-not-merge / do-not-auto-edit labels. Laptop primary at origin/main f7679c9: no wire-schema v1 record and no golden-vector corpus for the meta-protocol envelope (the postcard hits on main are the live pkgs/id encoding practice and old plans/spikes, exactly as the task 10211 triage recorded on 2026-11-05). No PR references issue id: design: postcard wire schema v1 (from #5674) #5808.
   • The gate, re-verified today: parent issue id: design: Wire format & transport adapters — postcard schema, non-UDP carriers, relays/proxies #5674 is still OPEN and its decision record has not landed — the 2026-11-08 done-task-only audit comment on issue id: design: Wire format & transport adapters — postcard schema, non-UDP carriers, relays/proxies #5674 confirms no merged change closes its full scope (PR docs(id): answers to Drew's 2026-11-04 ID program questions #5707 answered program questions only). Issue id: design: Wire format & transport adapters — postcard schema, non-UDP carriers, relays/proxies #5674's remainder is actively in flight right now: task 12244 triaged its remainder and handed it to the opencode claimant on 2026-11-08, and that ordering note names issue id: design: postcard wire schema v1 (from #5674) #5808 first among the decomposition children (id: design: postcard wire schema v1 (from #5674) #5808–id: design: security binding review across adapters & relays (from #5674) #5815).
   • Routing (no new packet, deliberately): a fix packet for issue id: design: postcard wire schema v1 (from #5674) #5808 claimed today could only re-discover this gate — the design cannot start until issue id: design: Wire format & transport adapters — postcard schema, non-UDP carriers, relays/proxies #5674's record lands, and the parent's own live route (task 12244 and its successors) is what unblocks it. To make the gate machine-visible instead of prose-only, the native blocked-by relation issue id: design: postcard wire schema v1 (from #5674) #5808 → issue id: design: Wire format & transport adapters — postcard schema, non-UDP carriers, relays/proxies #5674 was added and verified by readback this run (blocked_by: [#5674 open]). When issue id: design: Wire format & transport adapters — postcard schema, non-UDP carriers, relays/proxies #5674's decision record lands, issue id: design: postcard wire schema v1 (from #5674) #5808's design pass is the named next action and should be enqueued then.
   • Why this re-triage exists: prior triage task 10211 (2026-11-05) recorded the same gate but its queue task was marked done with no landed evidence, so issue queue: no reconciliation between open issues and queue coverage — non-creator filings never get queue items #7016's doctrine re-queued it; this note supersedes it with today's fresh verification and the native relation.
10. dezren39 commented on Nov 8, 2026

    

    dezren39

    on Nov 8, 2026

    MemberAuthor

    More actions

    🤖 Agent note · triaged — real and current, still parent-gated; blocked on #5674 rather than cycled again

    • Agent: OpenChamber scheduled lane (repo-worker-code, lane aea12c82-c5af-4e5a-9b66-f8e360f87177)
    • Session / ID: queue task 12892 (small_fix, repo-health-code-pilot)
    • Found via: atomic queue claim (re-queued by the 2026-11-07 done-task-only audit, tracking queue: no reconciliation between open issues and queue coverage — non-creator filings never get queue items #7016)

    Verdict. Same as the two triages before it, re-verified fresh at 2026-11-08T09:1xZ: this is a real design ticket whose deliverable does not exist yet, and whose own terms forbid writing it before the parent's decisions land.

    • Issue id: design: postcard wire schema v1 (from #5674) #5808 OPEN, no held / user/owned / do-not-merge / do-not-auto-edit labels; labels task/triage, app/id, status/triaged.
    • At origin/main 2b3ca04 there is no wire-schema v1 decision record and no golden-vector corpus for the meta-protocol envelope. The postcard hits under doc/ are the 2026-10/2026-11 design and spike material plus the live pkgs/id encoding practice (src/protocol.rs, src/meta_client.rs, src/peer_announcement.rs, src/match_audit.rs) — exactly what the 2026-11-05 triage recorded. No PR references id: design: postcard wire schema v1 (from #5674) #5808 (a --search 5808 PR query returns only unrelated id(attestation): compute_state never checks the two players are distinct — players=[A,A] attests Agreed from one key's single genesis + single claim #2863 duplicate-seat work).
    • Parent id: design: Wire format & transport adapters — postcard schema, non-UDP carriers, relays/proxies #5674 remains OPEN (queue/parked, work/assigned, status/parked) with no decision record landed. Its remainder is in the queue as task 12244, which is AT its attempt cap (3/3) and was handed to the opencode claimant 2026-11-08T06:00Z.

    Why this task is blocked and not re-enqueued. The acceptance for a re-triage says to convert a real issue into a fix packet — but this task already is the fix-class packet (repo-health:v1:fix:…:issue:5808), and a fresh one claimed today could only rediscover the same gate: the design cannot be written until #5674 records the wire-format family and transport-adapter decisions this ticket must not re-decide. Cycling the triage again is the loop this issue has already been through twice (tasks 10211, 12309). Blocking records the dependency honestly and keeps the real work queued for the moment the parent lands.

    Resume condition. #5674's decision record lands (task 12244 reaches a terminal state with the record merged), then write the postcard wire-schema v1 record here: envelope, discriminators, golden vectors, alternatives rejected, and the invariants implementation must hold — ratifying the existing pkgs/id postcard practice rather than inventing a new encoding.

    • Links: id: design: postcard wire schema v1 (from #5674) #5808 · parent id: design: Wire format & transport adapters — postcard schema, non-UDP carriers, relays/proxies #5674 (task 12244) · epic epic: ID platform expansion — apps, comms, currency, hosting, boards (2026-10-28 directive, LIST 6–15) #2056 · queue: no reconciliation between open issues and queue coverage — non-creator filings never get queue items #7016
    • Queue state: blocked, reason class dependency; no status/* / queue/* / work/* label is set — nothing looks in-flight.
11. dezren39

    added

    queue/blockedQueue or user escalation stateQueue or user escalation state

    status/blockedQueue or user escalation stateQueue or user escalation state

    and removed

    status/triagedTriaged and routed (see agent note)Triaged and routed (see agent note)

    on Nov 8, 2026