Gaps from the adversarial review of the workflow language, engine and canvas. Already tracked: archie-core-t19m.8 (test-run drafts), archie-core-t19m.9 (version history), archie-core-v6zm (run the shipped workflows end to end), archie-core-ctcr (label routing overrides), archie-core-afbk.3 (pr-review as steps).
Imported from beads archie-core-rcxb · GitHub #1269 · type epic · priority P1 · created 2026-10-05
Gaps from the adversarial review of the workflow language, engine and canvas. Already tracked: archie-core-t19m.8 (test-run drafts), archie-core-t19m.9 (version history), archie-core-v6zm (run the shipped workflows end to end), archie-core-ctcr (label routing overrides), archie-core-afbk.3 (pr-review as steps).
---
_Imported from beads `archie-core-rcxb` · GitHub #1269 · type epic · priority P1 · created 2026-10-05_
Triaged: no code change is needed for this. No code change: the issue states no gap.
What the body actually contains: one sentence ("Gaps from the adversarial review of the workflow language, engine and canvas") and a list of five items it says are Already tracked — archie-core-t19m.8 (test-run drafts), t19m.9 (version history), v6zm (run the shipped workflows end to end), ctcr (label routing overrides), afbk.3 (pr-review as steps). It names no defect, no wrong behaviour, no wanted capability, and no way to tell when it is done. The repo's own rule (CLAUDE.md, Issues) is that "an issue is the plan: what is wrong or wanted, and how you know it is done" — this one is a tracker pointer, not a plan.
The five referenced tickets are also not this issue's work, and several are already in the tree:
version history: CHANGELOG [1.56.0] — "The History tab lists a workflow's revisions, diffs them as YAML and on the canvas, restores one into the editor, and shows which revision each run used."
pr-review as steps: internal/domain/workflow/review_steps.go.
Only "run the shipped workflows end to end" was not checked.
No workflow fits. implement and tdd both need a stated behaviour; tdd needs an observable wrong behaviour to reproduce, and the body describes none. feasibility would mean writing a design document for "the gaps", and the gap list is not in the issue, so the document would be invented scope rather than a design for a request. If the epic does have a real backlog, the five bead IDs above are its only concrete content and each belongs to its own ticket.
For the operator (recommended answer first): close#1547 as a duplicate tracker, since everything concrete in it is tracked elsewhere and two of those children have shipped — the epic adds no scope of its own. If instead there is an unlisted set of review gaps this epic is meant to hold, paste that list into the issue with what is wrong and how you know it is fixed, and it can be triaged into feasibility (no settled design exists for most of the workflow-language surface) or tdd.
Separately verified, not part of this issue: docs/prds/event-automation.md is cited as the design authority in 8 places (external-agent-harness.md:73 and :89, ui/src/sources/SourcesPanel.vue:37, ui/src/captures/event-types.ts:2, migrations 0006–0009 header comments) but does not exist — docs/prds/ holds only self-development-loop.md and external-agent-harness.md. I did not touch it and it is a follow-up, not this issue's work.
Triaged: no code change is needed for this. No code change: the issue states no gap.
What the body actually contains: one sentence ("Gaps from the adversarial review of the workflow language, engine and canvas") and a list of five items it says are Already tracked — archie-core-t19m.8 (test-run drafts), t19m.9 (version history), v6zm (run the shipped workflows end to end), ctcr (label routing overrides), afbk.3 (pr-review as steps). It names no defect, no wrong behaviour, no wanted capability, and no way to tell when it is done. The repo's own rule (CLAUDE.md, Issues) is that "an issue is the plan: what is wrong or wanted, and how you know it is done" — this one is a tracker pointer, not a plan.
The five referenced tickets are also not this issue's work, and several are already in the tree:
- test-run drafts: internal/domain/workflow/test_run.go, test_run_test.go, workflow/task/task.go IsTestRun.
- version history: CHANGELOG [1.56.0] — "The History tab lists a workflow's revisions, diffs them as YAML and on the canvas, restores one into the editor, and shows which revision each run used."
- label routing: internal/domain/workflow/routing.go (workflowForLabels / IntakeRoutes).
- pr-review as steps: internal/domain/workflow/review_steps.go.
Only "run the shipped workflows end to end" was not checked.
No workflow fits. implement and tdd both need a stated behaviour; tdd needs an observable wrong behaviour to reproduce, and the body describes none. feasibility would mean writing a design document for "the gaps", and the gap list is not in the issue, so the document would be invented scope rather than a design for a request. If the epic does have a real backlog, the five bead IDs above are its only concrete content and each belongs to its own ticket.
For the operator (recommended answer first): close #1547 as a duplicate tracker, since everything concrete in it is tracked elsewhere and two of those children have shipped — the epic adds no scope of its own. If instead there is an unlisted set of review gaps this epic is meant to hold, paste that list into the issue with what is wrong and how you know it is fixed, and it can be triaged into feasibility (no settled design exists for most of the workflow-language surface) or tdd.
Separately verified, not part of this issue: docs/prds/event-automation.md is cited as the design authority in 8 places (external-agent-harness.md:73 and :89, ui/src/sources/SourcesPanel.vue:37, ui/src/captures/event-types.ts:2, migrations 0006–0009 header comments) but does not exist — docs/prds/ holds only self-development-loop.md and external-agent-harness.md. I did not touch it and it is a follow-up, not this issue's work.
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Gaps from the adversarial review of the workflow language, engine and canvas. Already tracked: archie-core-t19m.8 (test-run drafts), archie-core-t19m.9 (version history), archie-core-v6zm (run the shipped workflows end to end), archie-core-ctcr (label routing overrides), archie-core-afbk.3 (pr-review as steps).
Imported from beads
archie-core-rcxb· GitHub #1269 · type epic · priority P1 · created 2026-10-05Triaged: no code change is needed for this. No code change: the issue states no gap.
What the body actually contains: one sentence ("Gaps from the adversarial review of the workflow language, engine and canvas") and a list of five items it says are Already tracked — archie-core-t19m.8 (test-run drafts), t19m.9 (version history), v6zm (run the shipped workflows end to end), ctcr (label routing overrides), afbk.3 (pr-review as steps). It names no defect, no wrong behaviour, no wanted capability, and no way to tell when it is done. The repo's own rule (CLAUDE.md, Issues) is that "an issue is the plan: what is wrong or wanted, and how you know it is done" — this one is a tracker pointer, not a plan.
The five referenced tickets are also not this issue's work, and several are already in the tree:
Only "run the shipped workflows end to end" was not checked.
No workflow fits. implement and tdd both need a stated behaviour; tdd needs an observable wrong behaviour to reproduce, and the body describes none. feasibility would mean writing a design document for "the gaps", and the gap list is not in the issue, so the document would be invented scope rather than a design for a request. If the epic does have a real backlog, the five bead IDs above are its only concrete content and each belongs to its own ticket.
For the operator (recommended answer first): close #1547 as a duplicate tracker, since everything concrete in it is tracked elsewhere and two of those children have shipped — the epic adds no scope of its own. If instead there is an unlisted set of review gaps this epic is meant to hold, paste that list into the issue with what is wrong and how you know it is fixed, and it can be triaged into feasibility (no settled design exists for most of the workflow-language surface) or tdd.
Separately verified, not part of this issue: docs/prds/event-automation.md is cited as the design authority in 8 places (external-agent-harness.md:73 and :89, ui/src/sources/SourcesPanel.vue:37, ui/src/captures/event-types.ts:2, migrations 0006–0009 header comments) but does not exist — docs/prds/ holds only self-development-loop.md and external-agent-harness.md. I did not touch it and it is a follow-up, not this issue's work.