Skip to content

multi-agent: paired model-routing eval for retrieval, review and mechanical-edit roles #6901

Open
0 of 4 issues completed

Description

@kyle-sexton

Problem

No published measurement covers Claude 5.5 models on this repo's subagent roles: codebase retrieval, code review and verification, and bounded mechanical edits. Routing defaults in plugins/multi-agent/reference/defaults.yaml therefore rest on judgment. Haiku 5.5 may take no role until that role passes this eval.

Proposal

After the Haiku 5.5 documentation PRs land, build a paired eval comparing these setups on the same tasks from this repo:

  • Haiku 5.5 at high and xhigh
  • Haiku 5.5 with an Opus 5.5 advisor, using Anthropic's Haiku coding prompt
  • Sonnet 5.5 at low, medium and high
  • Opus 5.5 at low and medium

Measure:

  • pass rate, or verified findings
  • tokens per completed task (input, output and thinking)
  • advisor call rate
  • wall time

State no prices. Use a judge from outside the candidate set, or Opus with its self-preference noted.

Acceptance

  • An eval suite covers the retrieval, review and mechanical-edit roles
  • Results are recorded and the routing defaults are confirmed or changed
  • A role moves to Haiku only when it matches the current model's accuracy at no higher token cost per completed task

Activity

  1. added
    needs-triageNot yet classified. Floor until a type and one priority tier are set.
    on Oct 11, 2026
  2. github-actions commented on Oct 11, 2026

    @github-actions
    Contributor

    Proposed work class: none

    Thanks for the report. Automated triage handed this issue to a maintainer.

    Automated triage. A maintainer reviews it.

  3. added
    needs-humanHuman-in-the-loop required; autonomous sessions must not resolve items carrying this.
    on Oct 11, 2026
  4. added theissue type on Oct 12, 2026
  5. removed
    needs-triageNot yet classified. Floor until a type and one priority tier are set.
    on Oct 12, 2026
  6. removed
    needs-humanHuman-in-the-loop required; autonomous sessions must not resolve items carrying this.
    on Oct 25, 2026
  7. added
    needs-triageNot yet classified. Floor until a type and one priority tier are set.
    needs-humanHuman-in-the-loop required; autonomous sessions must not resolve items carrying this.
    on Oct 25, 2026
  8. melodic-automation-lanes commented on Oct 25, 2026

    @melodic-automation-lanes
    Contributor

    The triage verdict left this issue at status: needs-decision:

    • no Size: line
    • no Complexity: line
    • no Uncertainty: line
    • no Flags: line
    • the triage run escalated
    • held by needs-human
    • no lanes App verdict comment shows whether an earlier verdict was flagged
  9. added
    status: needs-decisionAwaiting a human or maintainer judgment call.
    and removed
    needs-triageNot yet classified. Floor until a type and one priority tier are set.
    on Oct 25, 2026
  10. kyle-sexton commented on Oct 25, 2026

    @kyle-sexton
    MemberAuthor

    Split by the owner's HITL session (2026-10-24, C12, Q93-Q95) into sub-issues. Slices with no blocker can run in parallel; the rest wait on their blocked-by edges. Intake triage sets each slice's fields, then admission takes it.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    needs-humanHuman-in-the-loop required; autonomous sessions must not resolve items carrying this.status: needs-decisionAwaiting a human or maintainer judgment call.

    Type

    Fields

    No fields configured for Task.

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions