Repository navigation
Closed
Labels
claude-triagedIssue has been triaged by the Claude Code triage routine. Remove to re-triage.Issue has been triaged by the Claude Code triage routine. Remove to re-triage.media-buyIssue concerns the media-buy protocol domainIssue concerns the media-buy protocol domainrfcProtocol change — auto-adds to roadmap boardProtocol change — auto-adds to roadmap boardspec / protocol
Milestone
Description
Problem
AdCP 3.2 has one kind of hold: finalizing a proposal reserves its inventory until expires_at. #7661 lets a seller disclose its hold policy before finalize. Scheduled inventory that depends on a person (creator posts, podcast host-reads, episode sponsorships) needs four things the binary hold can't express.
- Extension. Buyers often need more time for legal review or client sign-off. Today the only path is to let the hold lapse and finalize again, which can lose the slot. RFC: disclose proposal hold policy before finalization #7661 explicitly says re-finalizing is not an extension mechanism.
- Ranking. Sellers of scheduled media routinely give a first hold to one buyer and a second hold to another. The second hold converts if the first is released or lapses. With right of first refusal, when the second holder wants to commit, the first holder gets a deadline to accept or release.
- Holds on a specific installment. For installment products, the thing being held is a specific post date or episode. Buyers need to see which installments are held and until when.
- Holds waiting on a person. A creator marketplace can't commit until the creator agrees. Finalize can run async (
submitted), but there's no way to say "tentative hold, waiting on talent, expected by X".
Sketch for WG discussion
- Extend: a
refine_proposalsaction on a committed proposal that returns a newexpires_at, orhold_unavailable.hold_policy(RFC: disclose proposal hold policy before finalization #7661) declares the maximum number of extensions and the maximum total hold duration. - Rank:
hold_rankon a committed proposal (1 = first hold). The buyer is notified when its rank changes. When a lower-ranked holder asks to convert, the first holder gets arespond_bydeadline (right of first refusal). - Per-installment readback:
holds[]entries of{product_id | installment_ref, rank, expires_at}on the committed proposal. - Waiting on a person: reuse
expected_confirmation_at(Scheduled media: booking rounds and expected confirmation time #8010, PR feat(schema): add expected_confirmation_at to submitted media-buy envelopes #8016) on a submitted finalize.
Use cases
- A creator marketplace selling slots in creators' upcoming videos, where each slot needs the creator's yes.
- Podcast host-read and episode sponsorships.
- Linear and event sponsorships, where first and second holds are standard practice.
Metadata
Metadata
Assignees
Labels
claude-triagedIssue has been triaged by the Claude Code triage routine. Remove to re-triage.Issue has been triaged by the Claude Code triage routine. Remove to re-triage.media-buyIssue concerns the media-buy protocol domainIssue concerns the media-buy protocol domainrfcProtocol change — auto-adds to roadmap boardProtocol change — auto-adds to roadmap boardspec / protocol
Type
Projects
- StatusShow more project fieldsShipped
Activity
bokelley commentedon Oct 9, 2026
Triage
Classification: RFC
Bucket(s): spec / protocol, media-buy
Status: ready-for-human
Milestone: Spec Backlog
What the experts said:
extendmust produce a new committedproposal_id(same invariant asfinalize), not mutateexpires_atin place; item 1 also requires RFC: disclose proposal hold policy before finalization #7661 (hold_policy) as a prerequisite for normativemax_extensions/total_hold_durationconstraints; item 2 needs a newproposal.rank_changedwebhook event type and normative scoping ofholds[]to the authenticated buyer's own entries only (cross-buyer leakage risk).expected_confirmation_aton thesubmittedarm landed in 3.3 (PR feat(schema): add expected_confirmation_at to submitted media-buy envelopes #8016 merged). Items 1 and 3 solve real agentic gaps for scheduled media sellers. Item 2 (hold ranking / ROFR) carries high seller-side adoption friction — thehold_rankqueue + rank-change notification +respond_bydeadline is OMS-level state machine logic that most seller agents will route viaextinstead — and ROFR deadline enforcement puts AdCP in the middle of a commercial dispute if either party contests the clock.My take: Item 4 is implemented; items 1 and 3 are worth doing once the sketch corrections below are incorporated. Item 2 is the WG call: advisory readback (Option A) vs. normative ROFR obligation (Option B).
Decision memo (proposed — pending WG ratification)
Class: normative
Refs: #8086, #7661
Dissent: none yet
Item 4 — closed.
expected_confirmation_atonsubmittedfinalize arm covers the "waiting on talent" case. No additional surface needed.Item 3 — recommended. Add optional
holds[]to committed proposal response, scoped to the authenticated buyer's own entries only.Item 1 — recommended with correction. Add
extendaction torefine_proposals; it MUST produce a new committedproposal_id(same immutability pattern asfinalize,parent_proposal_idpointing back). Requires #7661 to land first.Item 2 — WG decision needed:
Option A — advisory
hold_rankreadback (no ROFR enforcement):{ "proposal_status": "committed", "hold_rank": 2 }No new webhook events. ROFR stays bilateral. Low seller adoption friction; populate when supported.
Option B — normative ROFR with
proposal.rank_changedwebhook:New webhook event type + deadline enforcement required. Sellers implement full hold-queue state machine.
Option A is the low-friction path; Option B provides richer buyer-side visibility but adds governance exposure and a new webhook primitive.
Triaged by Claude Code. Session: https://claude.ai/code/session_01VGH8NA4vmCT1YJrp1ja6BR
Generated by Claude Code
bokelley commentedon Oct 9, 2026
WG decision brief (3.3)
Agenda status: consent round. Objections by Fri 2026-10-16 23:59 UTC. This covers item 1 (hold extension) only. Items 2 and 3 are not in this slice, and item 4 was found not done:
expected_confirmation_atis absent from the compactrefine_proposalsenvelope, so it is tracked separately as slice #8086-A.Draft PR: #8097 (experimental
media_buy.hold_extension, changesetminor). Single adopter: Availnet, a creator marketplace.Recommended shape: a new
extend_holdaction onrefine_proposals. It targets a committed, unexpired proposal and forbids every revise field, likefinalize. A batch containing it contains onlyextend_holdentries and is atomic. Success returns the existingfinalizedoutcome with a new committed successor (parent_proposal_id= source, samecommercial_termsandterms_digest, laterexpires_at). Failure isunable/hold_unavailablewithbatch_abortedsiblings. Sellers declaremedia_buy.proposal_refinement.hold_extension: true; buyers MUST NOT sendextend_holdotherwise.expires_atin place? Recommend yes. A mutatedexpires_atwould break proposal immutability and theparent_proposal_idlineage.INVALID_STATE), and a failed extension leaves it and its hold untouched (yes / no)? Recommend yes. This avoids two live holds on one slot, and a failed extension never costs the buyer the hold it already has.hold_policyfields, leaving limits as seller policy expressed throughhold_unavailable, and keep RFC: disclose proposal hold policy before finalization #7661 in the backlog? Or pull RFC: disclose proposal hold policy before finalization #7661'smin_durationinto 3.3? Recommend yes, ship withouthold_policy. RFC: disclose proposal hold policy before finalization #7661 has five open decisions and would put the whole proposal at risk.Known interaction: under #7954
terms_only,extend_holdextends the acceptance window. #7954 excludeshold_unavailableunderterms_only, so the draft allows it for extension declines only; that needs reconciling when #7954 lands.Default if no objection by Fri 2026-10-16 23:59 UTC: the recommendation as drafted.
docs(media-buy): name Availnet as the hold_extension adopter
feat(media-buy): expected_confirmation_at on refine_proposals submitt…
3 remaining items
docs(media-buy): reconcile extend_hold with hold_policy and terms_only
bokelley commentedon Oct 12, 2026
Correction to the item 1 brief above
Question 3 in the brief above recommended shipping
extend_holdwithouthold_policyand leaving #7661 in the backlog. That has changed. #7661 has moved into 3.3 as #8140 (brief on #7661), and #8097 is now stacked on it. Question 3 now reads:hold_policy(yes / no)? Recommend yes:hold_policysaysextendable: true. Absent meansfalse, and the seller declines withhold_unavailable.expires_atmust be no later than the originalhold_placed_atplusmax_total_duration, counted across the whole chain of extensions.hold_policyandhold_placed_atforward unchanged.media_buy.proposal_refinement.hold_extension: trueremains the agent-level gate for sendingextend_holdat all.media_buy.hold_policykeep limits as seller policy expressed throughhold_unavailable.The known interaction with #7954 is now resolved.
extend_holdapplies only to committed proposals that hold inventory. Against a terms-only commitment, the seller fails the request at task level withINVALID_STATE, so terms-only sellers never returnhold_unavailable.Questions 1 and 2 and the default are unchanged. #8097 now targets #8140's branch, so its diff shows only the
extend_holdslice. It also adds compliance vectors for extension: within the ceiling, at the ceiling across two steps, past it,hold_placed_atcarried forward, and declines.bokelley commentedon Oct 14, 2026
Process update: this slice is merging to
mainnow as experimental, so it ships in the 3.3 beta and can be tried hands-on. After the beta, the training agent will implement it (#8181), and the working group will then review all the related slices together, end to end, against working behaviour (hold policy #8140,extend_hold#8097, response collections #8141, follower counts #8098).As a result, Friday's date no longer locks anything in. Objections and comments on the questions above are still welcome here at any time, and the experimental label means the shape can still change with notice.
feat(media-buy): experimental extend_hold action for refine_proposals (…
docs(media-buy): allow expected_confirmation_at on a submitted extend…
feat(media-buy): experimental advisory hold ranking (rankable, hold_r…
feat(media-buy): experimental per-installment hold readback (holds) (#…
bokelley commentedon Oct 17, 2026
Items 2 and 3 are merged to
mainas experimental slices. Both ship in the 3.3 beta. The WG reviews them end to end with the rest of the hold work after the training agent supports them (#8181).media_buy.hold_ranking, Option A (advisory).hold_policy.rankabledefaults tofalse. Without it,available: truestill means an exclusive, first-position hold.hold_rank(1= first). A live hold's position never gets worse, so a stalehold_rankis an upper bound on the buyer's real position.accept_proposalmay returnPRODUCT_UNAVAILABLEorsubmitted.media_buy.installment_holds.holds[](product_id,installment_ref, and a per-entryhold_rankunder a rankable policy) shows how the one all-or-none hold maps onto installments. It isn't a set of separate holds, so finalize stays atomic.expires_at.holds[]is required only when the allocated products declare specific installments.Accepted as recommended, and not in these slices: a buyer-side
require_first_positionopt-out on finalize, which can be added later without breaking anything.Merge origin/main into conductor/33-installments-8230
fix(schema): correct the holds/bought-set framing and bound the produ…
feat(media-buy): add installments to the canonical product and let a …
bokelley commentedon Oct 31, 2026
Shipped in 3.3 (experimental) first in 3.3.0-beta.0 via #8140 (
hold_policy), #8097 (extend_hold), #8255 (advisory hold ranking), #8257 (per-installmentholds[]readback) and #8104 (expected_confirmation_aton a submitted finalize). Decisions: #8086 (comment) (items 2 and 3) and #8086 (comment) (item 1).#8537 withdrew installment selection to 4.0 but left these features in place. The end-to-end WG review of the hold slices waits on the training agent and compliance work, which stay open as #8181 and #8182.