Skip to content

Decide fate of hand-created fixture issues in aspiregregator-squad-test #1768

Closed

Description

@bradygaster

The problem

In bradygaster/aspiregregator-squad-test, issues #9 (epic), #10, and #11 were hand-created by an agent — they were not produced by Squad decomposing anything.

They exist because PR #8's merge correctly emitted a no-op: issue #2 was flat, with no decomposable children. Rather than fix the fixture, issues were authored by hand to give the continuation path something to chew on.

That makes any continuation test run against them dishonest. It proves the dispatch mechanics fire, but it does not prove Squad can decompose an epic and then continue through its own children — which is the actual behavior under test.

The decision

(a) Keep them. They're adequate for testing the dispatch hop in isolation, and there's a real cost to rebuilding fixtures.

(b) Scrap and re-decompose. Delete #9/#10/#11, run /squad plan (or the full research → plan → implement path) against a genuine epic, and let Squad produce its own children. Slower, but the resulting test actually exercises the path end-to-end.

Recommendation

Lean (b) — a hand-built fixture can pass while the real path is broken, which is the specific failure mode that cost several rounds of debugging already.

Note on the newer fixture

A second fixture repo, bradygaster/aspiregregator-squad-e2e, is what #1767 targets. Decide whether:

  • -test gets retired entirely in favor of -e2e, or
  • both are kept with clearly documented, non-overlapping purposes

Two near-identically-named fixture repos with unclear roles is its own source of confusion.

Related

#1767 — the live e2e re-run that depends on having an honest fixture.

Activity

  1. bradygaster commented on Aug 20, 2026

    @bradygaster
    CollaboratorAuthor

    Flight sequencing note — revised 2026-08-20.

    Wave: 1, but cleanup/decision only — not a blocker for the continuation fix.

    Depends on: nothing in this stream.

    Blocks: nothing else in this stream.

    Safe to run in parallel with: all included issues. It does not touch bradygaster/squad files unless someone chooses to document fixture roles.

    Why / evidence: #1768 concerns the older bradygaster/aspiregregator-squad-test hand-created issues (#9/#10/#11). #1767 used the separate bradygaster/aspiregregator-squad-e2e fixture with epic #5 and real children #8/#9; the latest #1767 comment showed that honest fixture was good enough to surface #1772. Recommendation: do not spend time making -test the primary continuation fixture. Either retire it or explicitly document it as legacy/dispatch-hop-only so it cannot be mistaken for the end-to-end fixture again.

  2. added
    gh-awGitHub Agentic Workflows integration
    type:spikeResearch/investigation — produces a plan, not code
    triagedIssue has been triaged
    wave:1-nextWave 1 — immediate next sprint
    and removed
    questionCommunity question
    on Aug 20, 2026
  3. added
    squadSquad triage inbox — Lead will assign to a member
    on Aug 20, 2026
  4. bradygaster commented on Aug 20, 2026

    @bradygaster
    CollaboratorAuthor

    Flight triage note — scope expansion

    During gh-aw backlog triage I confirmed the same dishonest-fixture problem exists in a second fixture repo. This issue currently scopes only aspiregregator-squad-test (#9/#10/#11).

    Also in scope for this decision: bradygaster/aspiregregator-squad-e2e has leftover issues #12 and #14 ("Squad workflow dispatch missing command") produced by failed dispatch runs, not by Squad decomposing anything. Continuation tests that run against these are equally dishonest.

    Rather than open a near-duplicate issue, I'm folding both repos into this one decision. Recommendation (owner Sims): delete/close squad-test #9/#10/#11 and squad-e2e #12/#14, then regenerate continuation-test fixtures by having Squad actually decompose an epic, so the continuation path is exercised for real.

    This matters because #1772 and #1758 (the p0 end-to-end path defects) are validated against exactly these e2e runs — trustworthy fixtures are a prerequisite for trusting those fixes. Labeled wave:1-next for that reason.

    — Flight 🏗️

  5. bradygaster commented on Aug 20, 2026

    @bradygaster
    CollaboratorAuthor

    Closing — bookkeeping only. Decision was executed weeks ago.

    Sims confirmed: runs 32297494287–32297512862 in \�radygaster/aspiregregator-squad-test\ explicitly scrapped the hand-created fixture issues, using the rationale 'dishonest fixture.' squad-test issues #9, #10, and #11 are closed. \�spiregregator-squad-test\ is retired as the primary fixture. \�radygaster/aspiregregator-squad-e2e\ is the sole primary E2E fixture going forward — it contains Squad-decomposed issues (honest fixtures), not hand-created ones.

    This issue was never formally closed after the decision executed. Closing now as bookkeeping.

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

    gh-awGitHub Agentic Workflows integrationpriority:p2Next sprintsquadSquad triage inbox — Lead will assign to a membertriagedIssue has been triagedtype:spikeResearch/investigation — produces a plan, not codewave:1-nextWave 1 — immediate next sprint

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions