(feat-redraw)Integrate recruiter into the backend topology and SSE ecosystem#659
Merged
delthazor merged 4 commits intoJul 2, 2026
Merged
Conversation
…pology recruit_agents emits a STATE_PROGRESS_UPDATE that creates one node per discovered agent (carrying the OASF record and CID inline) plus an anchor node sharing the seeded "Agentic Recruiter" label, so the frontend collapses it onto the real node and re-points the discovered-agent edges there. The recruiter-service A2A hop drops the topology middleware so only the seeded recruiter and the discovered agents appear; the delegation middleware moves to dynamic_workflow_agent (its only remaining user). Co-authored-by: Cursor <cursoragent@cursor.com> Signed-off-by: pdobrei <pdobrei@cisco.com>
… renderer Discovered agents now arrive through the normal workflow-instance topology channel instead of a frontend overlay, so they anchor to the seeded "Agentic Recruiter" node, render with full OASF/directory parity, and are disposed of when the instance is torn down. The recruiter stream sends workflow_instance_id; the dynamic renderer maps inline oasf_record/agent_cid onto node data (OASF modal, directory link, target handle). Removes the discovery overlay (useMainAreaDiscoveryGraph, mergeDiscovery*, discoveryResponseEvent wiring), which mispositioned nodes, used a fragile recruiter-label match, and prevented disposal. Co-authored-by: Cursor <cursoragent@cursor.com> Signed-off-by: pdobrei <pdobrei@cisco.com>
…ance-id validator - extract _forward_status_update so recruit_agents and evaluate_agent share one intermediate A2A status-forwarding path instead of duplicating it - validate workflow instance ids via the public InstanceId schema model instead of importing the private _INSTANCE_ID_REGEX (recruiter_client + middleware) - add merge round-trip tests locking the inline oasf_record/agent_cid extras through create and partial-update so discovered-agent data is never dropped Co-authored-by: Cursor <cursoragent@cursor.com> Signed-off-by: pdobrei <pdobrei@cisco.com>
Resolve discovered-agent anchoring authoritatively in the store's serial merge worker instead of relying on frontend dedup. A new reconcile_event_node_identities transform collapses an incoming create-node whose stable_agent_id matches an existing node, drops the duplicate, and repoints edge endpoints to the existing runtime id. The recruiter anchor now carries the seeded recruiter's stable_agent_id so the graph stays a single continuous entity, and the normalized event is what gets dispatched over SSE. Co-authored-by: Cursor <cursoragent@cursor.com> Signed-off-by: pdobrei <pdobrei@cisco.com>
This was referenced Jun 23, 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.
Description
Recruiter as a pattern currently uses the recruiter agent purely from the UI side, and the discovered agents are added through an UI overlay which is not compatible with the backend-driven topology based dynamic graph rendering.
The proper way to do it is that the recruiter agent should add the discovered agent to the backend-stored topology (corresponding to the current workflow), and then SSE events should notify the UI that a new node needs to be rendered.
Topology graphs should be rendered dynamically based on backend provided data. This was partially done before, but was mostly reverted, due to stylistic and structural reasons, e.g.: specific icons were not rendered correctly, nodes not appearing as they should, etc.
This change aims to correct the recruiter pattern/workflow discrepancies by integrating the discovered agents into the existing ecosystem, so the automatic topology rendering can take care of the node representation correctly.
This PR is not a standalone change, it is a part of a chain of smaller changes reviewed independently.
The previous PR, this one immediately depends on:
#658
The dynamic graph rendering change, that this PR logically depends on:
#657
The follow-up PR is here:
#661
Other corresponding PRs:
#656
#663
Issue Link
No specific issue was opened for the change.
Type of Change
Checklist