ai-agent: serve the conversation view over a streamed HTTP route, not GraphQL - #167
Merged
Conversation
… GraphQL
The asz.view document is as large as the conversation, tens of megabytes
for a long one; the OAP now serves it from
GET /ai-agent/conversations/{conversation}/v1/view on the same server as
/graphql, streamed and compressed, under its own timeout. The
getConversationView query and the ConversationView type are removed; the
route is documented here, beside the two queries that remain.
This was referenced Sep 5, 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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Follow-up to #166.
The
asz.viewdocument is as large as the conversation it describes: a session of 136 MB of landed files renders to a 70 MB document. Measured on the OAP, building it takes about six seconds of storage reads plus five of fold and render, so as a GraphQL query it fails on the HTTP server's ten second default timeout, and even when it fits, the whole document has to exist as one JSON string on both sides.The OAP now serves the document from its own route on the same HTTP server as
/graphql, streamed as it is rendered, compressed when the client allows, under its own timeout:v1is the document version. The body is JSON, or YAML onAccept: application/yaml; 400 without a service, 404 when no round of the conversation is stored, 500 on a storage failure, each with{"errorReason": "..."}.This PR removes the
getConversationViewquery and theConversationViewtype, and documents the route in the file header beside the two queries that remain,listConversationsandgetConversationRawFiles.OAP side: apache/skywalking, branch
feat/ai-agent-conversation, unreleased (11.1.0), so nothing shipped depends on the removed query.