Review feedback for Activity Protocol Schema v5 proposal (issue #472)#473
Draft
Review feedback for Activity Protocol Schema v5 proposal (issue #472)#473
Conversation
Copilot
AI
changed the title
[WIP] Review proposal for Agents issue 472
Review feedback for Activity Protocol Schema v5 proposal (issue #472)
Mar 6, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Issue #472 proposes Activity Protocol v5: a minimal symmetric envelope with
payload-first multimodal content, unifiedrelatesTocorrelation, and typedentitiesfor metadata. No code changes are made — this PR documents the review findings.Critical Issues
relatesTo.stream(v5) vs.entities[streamInfo](Extend Activity Schema to Support Multimodal Interactions with Streaming - Final Proposal (Revision) #416);payload.contentType(v5) vs.value/valueType(Extend Activity Schema to Support Multimodal Interactions with Streaming - Final Proposal (Revision) #416) — incompatible designs, one already acceptedrelatesTosemantic collision: v3 A5200 explicitly prohibits intra-conversation references; v5 redefines it entirely for that purpose without retiring A5200namefield violates A6311: Examples useweather.getForecast/calendar.createEvent— not MIME media type format as requiredid/timestampownership flip: v3 requires channels to set these, SHOULD NOT for agents/clients; v5 makes them required for all senders — no migration rationale providedMajor Concerns
AXXXXmarkers — the current spec has hundreds; this is untraceable as a committee drafttyping,trace,invoke,handoff,endOfConversation,messageDelete,messageUpdate,messageReaction, etc. — no mapping or retirement guidancev: 5required with no negotiation strategy: Hard break for all existing implementations; no guidance on how dual-stack receivers gate behaviordelegationChain.proof.token: Raw tokens embedded in durable activity bodies expire before replay; potential logging/forwarding exposurepayloadconditionality undefined: "Required for content-bearing activities" — what makes an activity content-bearing? No normative definitionDesign-Level Feedback
toasobject | object[]: Union type forcesArray.isArray()guards everywhere in SDKs; prefer always-arraycausedByhas no normative semantics: Defined in the shape but Section 6 has no MUST/SHOULD for itdelivery.ttlMsunits: Milliseconds for TTL is error-prone; ISO 8601 duration (PT10M) is consistent with the rest of the speccommandResultsuccess/error envelope (§6.4.1): Labeled "interop guidance" but should be normative SHOULD to achieve its stated goalSuggested Before Committee Approval
payloadconditionality with normative requirementstoto always-array💬 We'd love your input! Share your thoughts on Copilot coding agent in our 2 minute survey.