mirror of
https://github.com/garrytan/gstack.git
synced 2026-10-03 01:46:55 +02:00
- C0/C1: the five never-green evals (skill-e2e-autoplan-chain and
skill-e2e-plan-{ceo,eng,design,devex}-finding-count) failed on harness and
budget, never on skill behavior; delete them, their touchfile/tier ids,
AUTOPLAN_CHAIN_BUDGET and the dedicated eighth periodic slice (--slices 7).
- C2: delete the helper groups whose only paid consumers were those files
(11 modules), trim claude-pty-runner and eng-seeded-coverage to the paid
closure, and delete the free replay tests whose assertions exercised only
that dead code (89 files, 135 orphaned fixtures). Blocks that used dead code
only as input for a live subject keep their assertions: the multiSelect
default moved to plan-review-decisions, runner PTY tests use inline caller
policies, and the timer-safe budget checks moved to eng-finding-retry-budget.
- The eight production-touching files stay except ceo-current-decision-record
(its template read only feeds the retired counter).
- CARVE_GUARDS.autoplan is behavioral 'none'; TODOS records the lost chain
and per-finding cadence coverage with their re-entry tests.
45 lines
2.5 KiB
JSON
45 lines
2.5 KiB
JSON
{
|
|
"$schema": "https://gstack.dev/schemas/section-manifest.json",
|
|
"skill": "autoplan",
|
|
"version": 1,
|
|
"note": "PASSIVE registry (v2 plan T9 / CM2). Fields are IDs, file paths, human titles, and human-readable trigger text ONLY. The skeleton's phase sequencing (Sequential Execution + the Phase 0 UI/DX scope detection) is the ONLY place that decides WHEN to read a section \u2014 Phase 2 and Phase 2.5 are conditional and their sections must NOT be read when their scope is absent; no paid eval checks the required section reads since the autoplan chain eval was retired (TODOS.md). No machine predicate here \u2014 see docs/designs/v2_PLAN.md:663.",
|
|
"sections": [
|
|
{
|
|
"id": "ceo-phase",
|
|
"file": "ceo-phase.md",
|
|
"title": "Phase 1: CEO review (strategy & scope) \u2014 override rules, dual voices, required outputs",
|
|
"trigger": "starting Phase 1 (CEO review \u2014 always runs, after the Phase 0.5 preflight)"
|
|
},
|
|
{
|
|
"id": "design-phase",
|
|
"file": "design-phase.md",
|
|
"title": "Phase 2: design review \u2014 override rules, dual voices, 7-pass checklist",
|
|
"trigger": "starting Phase 2 (design review \u2014 ONLY if UI scope was detected in Phase 0; skip the read entirely otherwise)"
|
|
},
|
|
{
|
|
"id": "eng-phase",
|
|
"file": "eng-phase.md",
|
|
"title": "Phase 3: eng review + dual voices \u2014 override rules, test review, required outputs",
|
|
"trigger": "starting Phase 3 (eng review \u2014 always runs, after all earlier applicable phases have closed)"
|
|
},
|
|
{
|
|
"id": "dx-phase",
|
|
"file": "dx-phase.md",
|
|
"title": "Phase 2.5: DX review \u2014 override rules, dual voices, 8-dimension scorecard",
|
|
"trigger": "starting Phase 2.5 (DX review \u2014 ONLY if developer-facing scope was detected in Phase 0; skip the read entirely otherwise)"
|
|
},
|
|
{
|
|
"id": "phase-close",
|
|
"file": "phase-close.md",
|
|
"title": "Close the current review phase \u2014 current readback, reconciliation and visible completion",
|
|
"trigger": "closing a review phase, after its reviews finish and before announcing completion or loading the next phase (read afresh at each exit)"
|
|
},
|
|
{
|
|
"id": "tasks-aggregator",
|
|
"file": "tasks-aggregator.md",
|
|
"title": "Phase 4: Implementation Tasks aggregator (builds $AGGREGATED_TASKS for the gate)",
|
|
"trigger": "presenting the Final Approval Gate (Phase 4) \u2014 the aggregator computes $AGGREGATED_TASKS that the gate message substitutes"
|
|
}
|
|
]
|
|
}
|