@1aboveio/engineering-skills 0.0.0-stage → 0.1.1
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.md +28 -2
- package/package.json +34 -4
- package/skills/better-review/LICENSE +3 -0
- package/skills/better-review/SKILL.md +249 -0
- package/skills/better-review/agents/openai.yaml +9 -0
- package/skills/better-review/references/domain-hazards.md +53 -0
- package/skills/better-review/references/passes.md +147 -0
- package/skills/better-review/references/result.md +151 -0
- package/skills/better-review/release-check.mjs +93 -0
- package/skills/better-review/scripts/pr-context.mjs +212 -0
- package/skills/better-review/scripts/pr-context.test.mjs +168 -0
- package/skills/better-review/scripts/rereview-scope.mjs +503 -0
- package/skills/better-review/scripts/rereview-scope.test.mjs +587 -0
- package/skills/better-review/scripts/scan-diff.mjs +365 -0
- package/skills/better-review/scripts/scan-diff.test.mjs +414 -0
- package/skills/better-review/scripts/vendor-sync.test.mjs +96 -0
- package/skills/e2e-test/LICENSE +3 -0
- package/skills/e2e-test/SKILL.md +98 -0
- package/skills/e2e-test/assets/ci-gates.cloudbuild.yaml +24 -0
- package/skills/e2e-test/references/authoring/auth-flows.md +159 -0
- package/skills/e2e-test/references/authoring/playwright-config.md +73 -0
- package/skills/e2e-test/references/authoring/playwright-patterns.md +144 -0
- package/skills/e2e-test/references/ci-integration.md +25 -0
- package/skills/e2e-test/references/ci-playwright-container.md +290 -0
- package/skills/e2e-test/references/presentation-sweep.md +104 -0
- package/skills/e2e-test/references/reviewing.md +41 -0
- package/skills/e2e-test/references/route-manifest.md +51 -0
- package/skills/e2e-test/release-check.mjs +93 -0
- package/skills/e2e-test/scripts/detect-routes-nextjs.mjs +200 -0
- package/skills/e2e-test/scripts/presentation-checks.mjs +171 -0
- package/skills/e2e-test/scripts/presentation-checks.test.mjs +106 -0
- package/skills/e2e-test/scripts/presentation-perceivability.mjs +179 -0
- package/skills/e2e-test/scripts/presentation-perceivability.test.mjs +117 -0
- package/skills/e2e-test/scripts/presentation-reachability.mjs +154 -0
- package/skills/e2e-test/scripts/presentation-reachability.test.mjs +69 -0
- package/skills/e2e-test/scripts/presentation-render-health.mjs +141 -0
- package/skills/e2e-test/scripts/presentation-render-health.test.mjs +98 -0
- package/skills/e2e-test/scripts/presentation-sweep.mjs +147 -0
- package/skills/e2e-test/scripts/presentation-sweep.test.mjs +71 -0
- package/skills/e2e-test/scripts/presentation-temporal.mjs +127 -0
- package/skills/e2e-test/scripts/presentation-temporal.test.mjs +59 -0
- package/skills/e2e-test/scripts/project-route-manifest.mjs +82 -0
- package/skills/e2e-test/scripts/project-route-manifest.test.mjs +76 -0
- package/skills/e2e-test/scripts/validate-manifest.mjs +106 -0
- package/skills/engineering-runtime/LICENSE +3 -0
- package/skills/engineering-runtime/SKILL.md +11 -0
- package/skills/engineering-runtime/agents/openai.yaml +6 -0
- package/skills/engineering-runtime/coherence/workflow.json +597 -0
- package/skills/engineering-runtime/engineering-runtime.test.mjs +68 -0
- package/skills/engineering-runtime/release-check.mjs +93 -0
- package/skills/engineering-runtime/scripts/exact-head-artifact.mjs +131 -0
- package/skills/engineering-runtime/scripts/head-check-set.mjs +398 -0
- package/skills/engineering-runtime/scripts/invocation-policy.mjs +187 -0
- package/skills/engineering-runtime/scripts/invocation-policy.test.mjs +95 -0
- package/skills/engineering-runtime/scripts/main-module.mjs +79 -0
- package/skills/engineering-runtime/scripts/main-module.test.mjs +288 -0
- package/skills/engineering-runtime/scripts/mergify-yaml.mjs +11 -0
- package/skills/engineering-runtime/scripts/package-lock.json +43 -0
- package/skills/engineering-runtime/scripts/package.json +10 -0
- package/skills/engineering-runtime/scripts/required-check-plan.mjs +223 -0
- package/skills/engineering-runtime/scripts/required-check-plan.test.mjs +96 -0
- package/skills/engineering-runtime/scripts/review-completion.mjs +56 -0
- package/skills/engineering-runtime/scripts/workflow-coherence.mjs +595 -0
- package/skills/engineering-runtime/scripts/workflow-policy.mjs +189 -0
- package/skills/engineering-runtime/scripts/yaml-mapping.mjs +18 -0
- package/skills/ensure-coverage/LICENSE +3 -0
- package/skills/ensure-coverage/SKILL.md +129 -0
- package/skills/ensure-coverage/references/breadth/coverage-ledger.md +92 -0
- package/skills/ensure-coverage/references/breadth/inventory-contract.md +85 -0
- package/skills/ensure-coverage/references/breadth/surface-baseline.md +44 -0
- package/skills/ensure-coverage/references/breadth/surface-discovery.md +16 -0
- package/skills/ensure-coverage/references/depth/characterization.md +68 -0
- package/skills/ensure-coverage/references/depth/coverage.config.example.json +19 -0
- package/skills/ensure-coverage/references/depth/grading.md +35 -0
- package/skills/ensure-coverage/references/depth/mock-policy.md +57 -0
- package/skills/ensure-coverage/references/depth/test-smells.md +39 -0
- package/skills/ensure-coverage/references/enforcement/ci-contract.md +44 -0
- package/skills/ensure-coverage/references/examples/coverage-ledger.md +110 -0
- package/skills/ensure-coverage/references/examples/refund-flow.md +35 -0
- package/skills/ensure-coverage/references/presentation/axis.md +64 -0
- package/skills/ensure-coverage/references/process/audit-mode.md +27 -0
- package/skills/ensure-coverage/references/process/output-template.md +139 -0
- package/skills/ensure-coverage/references/process/test-strategy.md +56 -0
- package/skills/ensure-coverage/release-check.mjs +93 -0
- package/skills/ensure-coverage/schemas/coverage-config.schema.json +67 -0
- package/skills/ensure-coverage/schemas/coverage-file.schema.json +140 -0
- package/skills/ensure-coverage/scripts/README.md +99 -0
- package/skills/ensure-coverage/scripts/adapters/nextjs-inventory.mjs +182 -0
- package/skills/ensure-coverage/scripts/adapters/nextjs-inventory.test.mjs +96 -0
- package/skills/ensure-coverage/scripts/check-quarantine-expiry.mjs +105 -0
- package/skills/ensure-coverage/scripts/check-quarantine-expiry.test.mjs +47 -0
- package/skills/ensure-coverage/scripts/ci-audit.mjs +137 -0
- package/skills/ensure-coverage/scripts/ci-audit.test.mjs +86 -0
- package/skills/ensure-coverage/scripts/coverage-ledger.mjs +674 -0
- package/skills/ensure-coverage/scripts/coverage-ledger.test.mjs +447 -0
- package/skills/ensure-coverage/scripts/design-parity.mjs +591 -0
- package/skills/ensure-coverage/scripts/design-parity.test.mjs +552 -0
- package/skills/ensure-coverage/scripts/lint-tests.mjs +282 -0
- package/skills/ensure-coverage/scripts/lint-tests.test.mjs +289 -0
- package/skills/ensure-coverage/scripts/mock-policy-config.mjs +176 -0
- package/skills/ensure-coverage/scripts/mock-policy-config.test.mjs +119 -0
- package/skills/ensure-coverage/scripts/package.json +19 -0
- package/skills/golden-path/LICENSE +3 -0
- package/skills/golden-path/SKILL.md +230 -0
- package/skills/golden-path/agents/openai.yaml +3 -0
- package/skills/golden-path/assets/golden-paths.example.json +57 -0
- package/skills/golden-path/references/contract.md +239 -0
- package/skills/golden-path/references/migration.md +111 -0
- package/skills/golden-path/release-check.mjs +93 -0
- package/skills/golden-path/scripts/golden-path-core.mjs +1779 -0
- package/skills/golden-path/scripts/golden-path.mjs +6 -0
- package/skills/golden-path/scripts/golden-path.test.mjs +1457 -0
- package/skills/harness-runtime/LICENSE +3 -0
- package/skills/harness-runtime/SKILL.md +19 -0
- package/skills/harness-runtime/agents/openai.yaml +6 -0
- package/skills/harness-runtime/bin/discover-models.mjs +4 -0
- package/skills/harness-runtime/bin/model-catalog.mjs +4 -0
- package/skills/harness-runtime/contracts.md +15 -0
- package/skills/harness-runtime/discover-models.mjs +392 -0
- package/skills/harness-runtime/discover-models.test.mjs +363 -0
- package/skills/harness-runtime/fixtures/native-question-schemas.json +33 -0
- package/skills/harness-runtime/fixtures/question-responses.json +54 -0
- package/skills/harness-runtime/harness-runtime.test.mjs +967 -0
- package/skills/harness-runtime/index.mjs +767 -0
- package/skills/harness-runtime/model-catalog-seed.test.mjs +829 -0
- package/skills/harness-runtime/model-catalog.mjs +787 -0
- package/skills/harness-runtime/model-catalog.test.mjs +366 -0
- package/skills/harness-runtime/native-question-contracts.md +37 -0
- package/skills/harness-runtime/references/model-catalog-seed.json +408 -0
- package/skills/harness-runtime/references/model-catalog.md +57 -0
- package/skills/harness-runtime/release-check.mjs +93 -0
- package/skills/resolve-issues/LICENSE +3 -0
- package/skills/resolve-issues/SKILL.md +250 -0
- package/skills/resolve-issues/agents/openai.yaml +9 -0
- package/skills/resolve-issues/plugin-hooks.json +17 -0
- package/skills/resolve-issues/references/canary.md +28 -0
- package/skills/resolve-issues/references/ci.md +105 -0
- package/skills/resolve-issues/references/done.md +25 -0
- package/skills/resolve-issues/references/dropbox.md +90 -0
- package/skills/resolve-issues/references/expire.md +318 -0
- package/skills/resolve-issues/references/exploration.md +202 -0
- package/skills/resolve-issues/references/implementation.md +303 -0
- package/skills/resolve-issues/references/preflight.md +445 -0
- package/skills/resolve-issues/references/review.md +228 -0
- package/skills/resolve-issues/references/shared-modules.md +44 -0
- package/skills/resolve-issues/references/task-plan.md +264 -0
- package/skills/resolve-issues/release-check.mjs +93 -0
- package/skills/resolve-issues/scripts/cleanup-runs.mjs +364 -0
- package/skills/resolve-issues/scripts/cleanup-runs.test.mjs +280 -0
- package/skills/resolve-issues/scripts/discover-models.mjs +9 -0
- package/skills/resolve-issues/scripts/dropbox.mjs +473 -0
- package/skills/resolve-issues/scripts/dropbox.test.mjs +346 -0
- package/skills/resolve-issues/scripts/findings-reuse.mjs +98 -0
- package/skills/resolve-issues/scripts/findings-reuse.test.mjs +88 -0
- package/skills/resolve-issues/scripts/github-reviews.mjs +384 -0
- package/skills/resolve-issues/scripts/github-reviews.test.mjs +264 -0
- package/skills/resolve-issues/scripts/install-hooks.mjs +135 -0
- package/skills/resolve-issues/scripts/install-hooks.test.mjs +79 -0
- package/skills/resolve-issues/scripts/issue-tree.mjs +162 -0
- package/skills/resolve-issues/scripts/issue-tree.test.mjs +107 -0
- package/skills/resolve-issues/scripts/manual-invocation.test.mjs +14 -0
- package/skills/resolve-issues/scripts/model-catalog.mjs +9 -0
- package/skills/resolve-issues/scripts/next-actions-capacity.test.mjs +149 -0
- package/skills/resolve-issues/scripts/next-actions.mjs +856 -0
- package/skills/resolve-issues/scripts/next-actions.test.mjs +915 -0
- package/skills/resolve-issues/scripts/plan-sync.mjs +269 -0
- package/skills/resolve-issues/scripts/plan-sync.test.mjs +156 -0
- package/skills/resolve-issues/scripts/preflight-models.mjs +736 -0
- package/skills/resolve-issues/scripts/preflight-models.test.mjs +448 -0
- package/skills/resolve-issues/scripts/ready-check.mjs +185 -0
- package/skills/resolve-issues/scripts/ready-check.test.mjs +217 -0
- package/skills/resolve-issues/scripts/rehydrate.mjs +98 -0
- package/skills/resolve-issues/scripts/rehydrate.test.mjs +91 -0
- package/skills/resolve-issues/scripts/review-verdict.mjs +509 -0
- package/skills/resolve-issues/scripts/review-verdict.test.mjs +1060 -0
- package/skills/resolve-issues/scripts/size-check.mjs +492 -0
- package/skills/resolve-issues/scripts/size-check.test.mjs +458 -0
- package/skills/resolve-release/LICENSE +3 -0
- package/skills/resolve-release/SKILL.md +215 -0
- package/skills/resolve-release/agents/openai.yaml +8 -0
- package/skills/resolve-release/flows/cloud-run-canary.md +13 -0
- package/skills/resolve-release/flows/cloud-run.md +7 -0
- package/skills/resolve-release/flows/root-tag-npm.md +13 -0
- package/skills/resolve-release/flows/workspace-npm.md +22 -0
- package/skills/resolve-release/references/canary.md +51 -0
- package/skills/resolve-release/references/candidate.md +41 -0
- package/skills/resolve-release/references/cut.md +131 -0
- package/skills/resolve-release/references/dark-launch.md +66 -0
- package/skills/resolve-release/references/flows.md +150 -0
- package/skills/resolve-release/references/preflight.md +154 -0
- package/skills/resolve-release/references/prepared-metadata.md +260 -0
- package/skills/resolve-release/references/promotion.md +82 -0
- package/skills/resolve-release/references/publish.md +232 -0
- package/skills/resolve-release/references/workspace-npm.md +103 -0
- package/skills/resolve-release/release-check.mjs +93 -0
- package/skills/resolve-release/scripts/apply.mjs +212 -0
- package/skills/resolve-release/scripts/apply.test.mjs +99 -0
- package/skills/resolve-release/scripts/dark-flags.mjs +134 -0
- package/skills/resolve-release/scripts/dark-flags.test.mjs +100 -0
- package/skills/resolve-release/scripts/flow.mjs +500 -0
- package/skills/resolve-release/scripts/flow.test.mjs +340 -0
- package/skills/resolve-release/scripts/green-head.mjs +231 -0
- package/skills/resolve-release/scripts/green-head.test.mjs +101 -0
- package/skills/resolve-release/scripts/npm-version-absence.mjs +138 -0
- package/skills/resolve-release/scripts/npm-version-absence.test.mjs +426 -0
- package/skills/resolve-release/scripts/plan-npm.test.mjs +659 -0
- package/skills/resolve-release/scripts/plan-toml-boundary.test.mjs +317 -0
- package/skills/resolve-release/scripts/plan-uv.test.mjs +291 -0
- package/skills/resolve-release/scripts/plan.mjs +806 -0
- package/skills/resolve-release/scripts/plan.test.mjs +428 -0
- package/skills/rush-issues/LICENSE +3 -0
- package/skills/rush-issues/SKILL.md +58 -0
- package/skills/rush-issues/agents/openai.yaml +8 -0
- package/skills/rush-issues/rehydrate.mjs +10 -0
- package/skills/rush-issues/release-check.mjs +93 -0
- package/skills/rush-release/LICENSE +3 -0
- package/skills/rush-release/SKILL.md +35 -0
- package/skills/rush-release/agents/openai.yaml +8 -0
- package/skills/rush-release/release-check.mjs +93 -0
- package/skills/test-quality/LICENSE +3 -0
- package/skills/test-quality/SKILL.md +79 -0
- package/skills/test-quality/references/audit.md +80 -0
- package/skills/test-quality/references/lint.md +24 -0
- package/skills/test-quality/references/principles.md +143 -0
- package/skills/test-quality/references/testing-conventions.md +9 -0
- package/skills/test-quality/release-check.mjs +93 -0
- package/skills/test-quality/scripts/lint-tests.mjs +574 -0
- package/skills/test-quality/scripts/lint-tests.test.mjs +591 -0
- package/skills/test-quality/scripts/mock-policy-config.mjs +176 -0
- package/skills/test-quality/scripts/mock-policy-config.test.mjs +119 -0
- package/skills/test-quality/scripts/persist-conventions-core.mjs +117 -0
- package/skills/test-quality/scripts/persist-conventions.mjs +17 -0
- package/skills/test-quality/scripts/persist-conventions.test.mjs +104 -0
|
@@ -0,0 +1,445 @@
|
|
|
1
|
+
# Preflight
|
|
2
|
+
|
|
3
|
+
Run these steps in order, in every run, including a one-unit run. Ask nothing
|
|
4
|
+
until step 6.
|
|
5
|
+
|
|
6
|
+
## 0. Ready label
|
|
7
|
+
|
|
8
|
+
**One run per spec.** Before anything else, find the spec's run:
|
|
9
|
+
|
|
10
|
+
```bash
|
|
11
|
+
node <skill-base-dir>/scripts/plan-sync.mjs --find-run --runs .resolve-issues/rush --issue <owner/name#123>
|
|
12
|
+
```
|
|
13
|
+
|
|
14
|
+
When it prints a directory, resume that run (`next-actions.mjs --run <dir> --resume` with
|
|
15
|
+
[fresh capacity](implementation.md#live-capacity))
|
|
16
|
+
instead of starting another; its plan picks up whatever the tree gained
|
|
17
|
+
since. It exits 0 whether or not it found a run: no run prints nothing on stdout
|
|
18
|
+
(`--json`: `{"found":false}`). Example: `/rush-issues #12` finds the run whose saved tree holds
|
|
19
|
+
#12 and resumes it, never a second run beside it.
|
|
20
|
+
|
|
21
|
+
For GitHub issues, first run, with every issue the user named:
|
|
22
|
+
|
|
23
|
+
```bash
|
|
24
|
+
node <skill-base-dir>/scripts/ready-check.mjs --issue <owner/name#123> [--issue ...]
|
|
25
|
+
```
|
|
26
|
+
|
|
27
|
+
The script exits 0 whenever it ran (2 if it could not); read `outcome` in its JSON.
|
|
28
|
+
|
|
29
|
+
- **`ready`:** continue. An open issue is work, whatever its sub-issues
|
|
30
|
+
say: each `rollups` entry (open, every sub-issue closed) is a unit; the
|
|
31
|
+
explorer records what is left, or that nothing is, and then the handback
|
|
32
|
+
says "close #N". Example: held #40's filed prerequisite #41 closed;
|
|
33
|
+
#40 is a rollup and stays a unit.
|
|
34
|
+
- **`unlabeled`:** continue as on `ready`, and record the `missing` issues for
|
|
35
|
+
the board and the step 6 ready-label question; shortcut mode warns instead of asking
|
|
36
|
+
([Shortcut mode](#shortcut-mode)). Never add the label yourself.
|
|
37
|
+
Example: `/resolve-issues #2975-2977` with none labeled explores all three
|
|
38
|
+
and asks at step 6; `/rush-issues` warns and implements them.
|
|
39
|
+
- **`unreadable`:** end the run. Report the `unreadable` issues; nothing can be
|
|
40
|
+
planned from an issue that cannot be read.
|
|
41
|
+
- **Exit 2** (the script could not run): fix the issue reference and rerun.
|
|
42
|
+
|
|
43
|
+
Skip this step for a local or pasted spec. Count only GitHub's native
|
|
44
|
+
sub-issues, never a checklist.
|
|
45
|
+
|
|
46
|
+
### Compaction hook
|
|
47
|
+
|
|
48
|
+
**A compaction summary keeps the run's state and drops the skill's rules, so
|
|
49
|
+
a hook makes you read the skills again after every compaction.** Check it:
|
|
50
|
+
|
|
51
|
+
```bash
|
|
52
|
+
node <skill-base-dir>/scripts/install-hooks.mjs --check
|
|
53
|
+
```
|
|
54
|
+
|
|
55
|
+
It exits 0; when it prints `missing: ...`, ask at step 6 (shortcut mode: warn on the board and carry on).
|
|
56
|
+
On yes, run it without `--check`: it adds a `SessionStart` hook with matcher
|
|
57
|
+
`compact` that runs `rehydrate.mjs`, to `~/.claude/settings.json` and
|
|
58
|
+
`~/.codex/hooks.json`, pointing at the copy of `rehydrate.mjs` it runs
|
|
59
|
+
from, so run the installed skill's copy, never a checkout or worktree that
|
|
60
|
+
may be removed. Claude Code's docs say it reloads hook settings during a
|
|
61
|
+
session; Codex may ask to trust the hook at its next start. With the
|
|
62
|
+
`engineering-workflow` plugin enabled, Claude Code already has the hook and
|
|
63
|
+
the script leaves its settings alone. Pi gets the re-read from its workflow
|
|
64
|
+
extension. Never install it without the human's yes: it changes their
|
|
65
|
+
machine-wide settings.
|
|
66
|
+
|
|
67
|
+
Then run `dropbox.mjs pending --runs .resolve-issues/rush`. When `runs` is not empty, show
|
|
68
|
+
each earlier run's unposted entries on the board, with the command that
|
|
69
|
+
posts them; they join this run's end-of-run review
|
|
70
|
+
([dropbox.md](dropbox.md#post-at-the-end)).
|
|
71
|
+
|
|
72
|
+
## 1. Model slots
|
|
73
|
+
|
|
74
|
+
Run the fast path and record its fills by exact id and effort, using the
|
|
75
|
+
ladders in [shared-modules.md](shared-modules.md#model-catalog). Never give
|
|
76
|
+
the reviewer the implementer's id, at any effort. Let risk raise the tier and
|
|
77
|
+
add the repository's deterministic domain checks, never change the workflow.
|
|
78
|
+
|
|
79
|
+
Pass a pinned implementer (`implementer=<id[@effort]>` in the invocation) as
|
|
80
|
+
`--implementer`; when the result is `ok: false` with code `pinned-implementer-unavailable`, end the run as "pinned implementer unavailable"
|
|
81
|
+
and never substitute another model.
|
|
82
|
+
Example: `/rush-issues 1aboveio/ai#1545 implementer=gpt-6-luna@high` with no
|
|
83
|
+
`gpt-6-luna` discovered stops at preflight and starts nothing.
|
|
84
|
+
|
|
85
|
+
## 2. Provisional target
|
|
86
|
+
|
|
87
|
+
Fetch remote heads; record the default branch as `targetBranch`.
|
|
88
|
+
|
|
89
|
+
## 3. Plan draft
|
|
90
|
+
|
|
91
|
+
Follow [task-plan.md](task-plan.md#initialize-the-run) up to creating
|
|
92
|
+
branches. Show the first board.
|
|
93
|
+
|
|
94
|
+
**Plan only what a run can finish.** A leaf issue whose deliverable is a
|
|
95
|
+
release, a physical-device campaign, a human approval or sign-off, or a
|
|
96
|
+
live-environment action is not a unit: record it in the plan's `excluded`
|
|
97
|
+
with the reason, so the issue-tree sync never adds it, and name it in the
|
|
98
|
+
handback as a remaining obligation of its own issue.
|
|
99
|
+
Example: epic #12's #20 (signed release and go/no-go) and #21
|
|
100
|
+
(physical acceptance) stay out of the plan; the run ends when its code units
|
|
101
|
+
land instead of waiting on them.
|
|
102
|
+
|
|
103
|
+
## 4. Explore
|
|
104
|
+
|
|
105
|
+
Run [exploration.md](exploration.md) on the provisional target, with any
|
|
106
|
+
prior findings for the same spec. On tool failures or a size-check `exploration-unfinished` verdict,
|
|
107
|
+
continue the explorer with the errors. Add the links the findings show.
|
|
108
|
+
|
|
109
|
+
## 5. Size check
|
|
110
|
+
|
|
111
|
+
```bash
|
|
112
|
+
node <skill-base-dir>/scripts/size-check.mjs \
|
|
113
|
+
--plan <run>/task-plan.json --findings <run>/exploration/findings.json
|
|
114
|
+
```
|
|
115
|
+
|
|
116
|
+
The script exits 0 and prints `verdict` (`ok`, `warn`, `hold`, `stop`, or `exploration-unfinished`); 2 means a bad plan or command.
|
|
117
|
+
A failed measure tagged with a unit holds that unit, and its dependents wait not started; an
|
|
118
|
+
untagged `plan` failure stops the run until its question is answered: at the
|
|
119
|
+
preflight approval in `resolve-issues`; in `rush-issues` or after preflight,
|
|
120
|
+
by row 1 of [Never block on the human](#never-block-on-the-human).
|
|
121
|
+
|
|
122
|
+
| Consideration | Measure | Warn | Hold or stop |
|
|
123
|
+
|---|---|---|---|
|
|
124
|
+
| scope | Behaviors | 2-3 | 4+: hold |
|
|
125
|
+
| scope | Missing prerequisite | any: the unit builds it, counted in its lines | |
|
|
126
|
+
| scope | Estimated lines x 1.3 | 2,501-5,000 | over 5,000: hold |
|
|
127
|
+
| proof | `browser`/`emulator` proof with no working end-to-end suite | any: the unit builds the journey it needs, counted in its lines | |
|
|
128
|
+
| proof | `emulator` proof across 3+ layers | any | |
|
|
129
|
+
| proof | `proof.physical` evidence | any: one pending release-gate record | |
|
|
130
|
+
| plan | Open `product`/`scope`/`approval` question for a unit | | any: hold that unit |
|
|
131
|
+
| plan | Blocking question with no unit, or an unknown unit | | stop |
|
|
132
|
+
| plan | Dependency chain | 6+ units: PRs land in sequence | |
|
|
133
|
+
|
|
134
|
+
| `verdict` | Do |
|
|
135
|
+
|---|---|
|
|
136
|
+
| `ok`, `warn` | Continue |
|
|
137
|
+
| `hold` | Hold the `held` units (their dependents wait, not started); start the `runnable` units |
|
|
138
|
+
| `stop` | Run-wide failure or every unit held; draft the held units' splits |
|
|
139
|
+
| `exploration-unfinished` | Continue the explorer with the error |
|
|
140
|
+
| exit 2 | Fix the plan or command |
|
|
141
|
+
|
|
142
|
+
Only scope and open questions hold a unit; proof only warns. Add each
|
|
143
|
+
`handback` item, including `verification`, as a handback step. A unit with
|
|
144
|
+
no working end-to-end suite builds the journey its proof needs inside the
|
|
145
|
+
unit ([expire.md](expire.md#proof-stays-with-its-unit)); never hold units
|
|
146
|
+
behind a suite.
|
|
147
|
+
|
|
148
|
+
### Splits
|
|
149
|
+
|
|
150
|
+
**Draft a split for each unit held on `scope` with no open
|
|
151
|
+
`product`, `scope`, or `approval` question; never split a unit with one.**
|
|
152
|
+
Example: #30, held at 4,300 x 1.3 lines, gets a split; a paused issue sized
|
|
153
|
+
at 6,000 lines is held by its `approval` question and gets none.
|
|
154
|
+
|
|
155
|
+
- Write it to `<run>/splits/<unit>.md`, one section per slice in the
|
|
156
|
+
`to-tickets` shape: `Parent`, `What to build`, `Acceptance criteria` (the
|
|
157
|
+
held issue's criteria for it), `Blocked by` (only slices whose change it
|
|
158
|
+
needs, [exploration.md](exploration.md) step 8).
|
|
159
|
+
- Start from the size check's `proposedSplit`. When it is `null`, cut the
|
|
160
|
+
slices by `to-tickets`' rules: read the first that exists of
|
|
161
|
+
`$(realpath <skill-base-dir>)/../to-tickets/SKILL.md`,
|
|
162
|
+
`~/.agents/skills/to-tickets/SKILL.md`, and
|
|
163
|
+
`~/.claude/skills/to-tickets/SKILL.md`, and follow its step 3 only
|
|
164
|
+
(vertical slices, prefactor first, blocking edges); never its quiz or
|
|
165
|
+
publish steps. When none exists, cut end-to-end slices, prefactors
|
|
166
|
+
first.
|
|
167
|
+
- Cut for width: most slices should be able to start together. Add a
|
|
168
|
+
prefactor only when two or more slices need its code and the first of
|
|
169
|
+
them cannot build it within the line limit; never chain slices that only
|
|
170
|
+
share a theme.
|
|
171
|
+
Example: four proof slices share a device helper; the first builds it
|
|
172
|
+
and the other three depend on it, so after the first, three run at once.
|
|
173
|
+
- Make each slice safe to merge alone; never leave a later slice to make
|
|
174
|
+
it correct or safe. Example: a write path and the check that refuses its
|
|
175
|
+
bad input go in one slice; split apart, the write path merges unguarded.
|
|
176
|
+
- Put proof with the change it proves while that change is unmerged:
|
|
177
|
+
never cut a slice that only adds tests, verification, or docs for code
|
|
178
|
+
another slice of the same split merges. Example: "run the full suite
|
|
179
|
+
against the new schema" belongs to the schema change's slice.
|
|
180
|
+
Proof of code already merged on the target is a legitimate slice of its
|
|
181
|
+
own. Example: an emulator campaign that proves merged features splits
|
|
182
|
+
along its `proof.groups` into a setup prefactor and group slices.
|
|
183
|
+
- Give each of the held issue's acceptance criteria to exactly one slice;
|
|
184
|
+
keep together criteria that the two rules above join, and merge
|
|
185
|
+
`proposedSplit` slices they join. When they join everything into one
|
|
186
|
+
slice, draft no split: recount the unit's behaviors by
|
|
187
|
+
[exploration.md](exploration.md) step 3, rerun the size check, and hand
|
|
188
|
+
the unit back if it still holds. Cut the fewest slices that fit: each
|
|
189
|
+
up to 3 behaviors and under 3,800 estimated lines (before x1.3); never
|
|
190
|
+
4+ behaviors or 3,800+ estimated lines. Every slice costs a PR, a
|
|
191
|
+
review, a queue turn, and often an app build and journey, while two or
|
|
192
|
+
three behaviors merge as fast as one. Example: a held 4-behavior unit
|
|
193
|
+
splits in two, not four.
|
|
194
|
+
- Note its `physical` evidence there as pending at the release gate.
|
|
195
|
+
- Draft a prefactor shared by held units once and name it in the
|
|
196
|
+
`Blocked by` of only the slices that use it.
|
|
197
|
+
- Record the size check's `verification` record once, as pending, on the
|
|
198
|
+
issue that owns the release gate (or in the handback when none exists);
|
|
199
|
+
never one issue per unit.
|
|
200
|
+
- Never edit or close the held issue. The only issues a run files are
|
|
201
|
+
splits ([File a split](#file-a-split)), the [dropbox](dropbox.md)'s issues, infrastructure
|
|
202
|
+
blockers ([expire.md](expire.md#infrastructure-failures)), and the one record of a
|
|
203
|
+
remaining obligation in its owning issue.
|
|
204
|
+
|
|
205
|
+
### Filed text
|
|
206
|
+
|
|
207
|
+
**Text a run files, posts, or edits in carries the parent's acceptance
|
|
208
|
+
criteria and the facts, and never adds a restriction.** That covers splits,
|
|
209
|
+
blocker issues, dropbox issues, status and handback comments, and edits to
|
|
210
|
+
the spec's or a unit's body. It goes out under the human's account, where
|
|
211
|
+
later runs cannot tell it from the human's own words, so it never adds a
|
|
212
|
+
prohibition (only [standing orders](task-plan.md#standing-orders) constrain a run), an authorization or approval step, an ownership or custody
|
|
213
|
+
requirement, or a new acceptance criterion. A human-only blocker is stated
|
|
214
|
+
as the one decision needed, not as a standing rule.
|
|
215
|
+
Example: "Emulator boot fails: `emulator -avd ci` exits 1 with `KVM not
|
|
216
|
+
available`; tried a cold boot and a fresh AVD", never "a maintainer must
|
|
217
|
+
authorize emulator use".
|
|
218
|
+
|
|
219
|
+
### Put every filed issue in the tree
|
|
220
|
+
|
|
221
|
+
**An issue the spec still needs goes into the spec's issue tree, with a
|
|
222
|
+
native `blocked_by` link on each issue that waits for it; the
|
|
223
|
+
[issue-tree sync](task-plan.md#issue-tree-sync) adds its unit.** Never wire
|
|
224
|
+
it into the plan by hand, and never leave it outside the tree. Example: a
|
|
225
|
+
prerequisite for held #40 is filed as a sub-issue of #40's parent #12, and
|
|
226
|
+
#40 gets `blocked_by` it; the next `next-actions.mjs --sync` prints
|
|
227
|
+
`add-unit` for it and `set-deps` for #40.
|
|
228
|
+
|
|
229
|
+
- A prerequisite (prefactor, infrastructure blocker) is a
|
|
230
|
+
sibling of the issue that waits, under that issue's parent, never its
|
|
231
|
+
sub-issue: a closed sub-issue would leave the waiting issue looking done.
|
|
232
|
+
- Put `blocked_by` on each issue that waits, never on an issue with open
|
|
233
|
+
sub-issues: GitHub shows it holding all of them while the run ignores it, so the sync prints
|
|
234
|
+
`move-link` until it moves. Example: a prerequisite two of five slices
|
|
235
|
+
use gets `blocked_by` on those two, not on their parent.
|
|
236
|
+
- A slice is a sub-issue of the held issue only when the slices together
|
|
237
|
+
carry every one of its acceptance criteria.
|
|
238
|
+
- Dropbox issues stay out of the tree: they block nothing.
|
|
239
|
+
|
|
240
|
+
After filing, run `next-actions.mjs --run <run> --sync` with
|
|
241
|
+
[fresh capacity](implementation.md#live-capacity) and apply what it
|
|
242
|
+
prints.
|
|
243
|
+
|
|
244
|
+
### File a split
|
|
245
|
+
|
|
246
|
+
**File a split only when the approval says so; shortcut mode always does.**
|
|
247
|
+
Example: `/resolve-issues` answered "run the rest" files nothing;
|
|
248
|
+
`/rush-issues` files #30's split.
|
|
249
|
+
|
|
250
|
+
1. List the held issue's sub-issues and its parent's first; reuse a slice
|
|
251
|
+
already filed under the same title and file only the rest.
|
|
252
|
+
2. Create one issue per slice, blockers first, labeled `ready-for-agent`,
|
|
253
|
+
placed by [Put every filed issue in the tree](#put-every-filed-issue-in-the-tree):
|
|
254
|
+
a prerequisite slice beside the held issue, with `blocked_by` on the
|
|
255
|
+
issues that use it (the held issue while it is one unit; once it has
|
|
256
|
+
slices, only the slices that use it); behavior slices under the
|
|
257
|
+
held issue only when they carry all of its criteria. File a prefactor
|
|
258
|
+
shared by held units once, beside the first held issue.
|
|
259
|
+
Slice text follows [Filed text](#filed-text).
|
|
260
|
+
Example: #30's split files "Profile read through the authenticated
|
|
261
|
+
handoff" and "Profile edit and common sign-out", blocked by the first,
|
|
262
|
+
both sub-issues of #30 because together they carry its criteria.
|
|
263
|
+
3. Run `next-actions.mjs --run <run> --sync` with
|
|
264
|
+
[fresh capacity](implementation.md#live-capacity) and apply its `add-unit`,
|
|
265
|
+
`set-deps`, and `drop-unit` lines: a held issue split into sub-issues
|
|
266
|
+
becomes a container and its unit drops; a held issue that only gained
|
|
267
|
+
prerequisites keeps its unit, now depending on them.
|
|
268
|
+
Example: #30 splits into #50 and #51, so the sync adds both, drops 30,
|
|
269
|
+
and unit 34, `blocked_by` #30, now waits on 50 and 51.
|
|
270
|
+
4. Continue the explorer for the new units only; rerun the size check.
|
|
271
|
+
5. Split a still-held slice again only while splitting makes progress: its
|
|
272
|
+
largest slice is smaller than the one before.
|
|
273
|
+
Example: a 7,287-line unit split into 3,455 and 2,150 lines; a held
|
|
274
|
+
3,455-line slice may split again, but not into one still sized 3,455.
|
|
275
|
+
6. **A held unit with no drafted split, or one no drafted split shrinks,
|
|
276
|
+
gets your recommended split.** Draft it in `<run>/splits/<unit>.md` as
|
|
277
|
+
to-tickets slices that together carry every acceptance criterion, each
|
|
278
|
+
smaller than the unit. Only when no cut does that (one behavior that
|
|
279
|
+
cannot be divided), recommend running it whole and set the unit's
|
|
280
|
+
`sizeCheck.runWhole: true` once accepted. `resolve-issues` hands back
|
|
281
|
+
the recommendation; `rush-issues` takes it ([Shortcut mode](#shortcut-mode)).
|
|
282
|
+
Example: a 9,386-line emulator campaign with no proof groups recorded
|
|
283
|
+
splits by feature family: shared setup first, then proof slices that
|
|
284
|
+
pack whole families up to the slice limits, each blocked by the setup.
|
|
285
|
+
|
|
286
|
+
## 6. Approval
|
|
287
|
+
|
|
288
|
+
Show the board: units, delivery mode, verdict and reasons by consideration,
|
|
289
|
+
each unit's behaviors, missing prerequisites, proof and suite, environment
|
|
290
|
+
checks, held units with splits, the pending device record, the
|
|
291
|
+
issues missing `ready-for-agent`, and handback steps. Ask
|
|
292
|
+
in one interaction, with the native single-choice tool when there is one:
|
|
293
|
+
|
|
294
|
+
0. **Ready label** (only after step 0 `unlabeled`): list the `missing` issues,
|
|
295
|
+
the ones the human named (`parent: null`) apart from sub-issues under
|
|
296
|
+
them. Options: proceed (recommended), or stop and start nothing. On
|
|
297
|
+
proceed, record it in `decisions`; on stop, end the run with the list.
|
|
298
|
+
**Compaction hook** (only when `install-hooks.mjs --check` printed `missing:`):
|
|
299
|
+
install it (recommended), or run without it.
|
|
300
|
+
1. **Implementer:** the payload's `options`, recommended first, each with its
|
|
301
|
+
reviewer. A pinned implementer is already decided: show it, don't ask.
|
|
302
|
+
2. **Target branch:** up to four fetched heads, provisional first.
|
|
303
|
+
3. **Merge mode:** `autonomous` (recommended): merge each
|
|
304
|
+
reviewed, green PR, fixing blockers outside the plan
|
|
305
|
+
([expire.md](expire.md#infrastructure-failures)); or
|
|
306
|
+
`supervised`: hand each back merge-ready.
|
|
307
|
+
4. **Scheduler mode:** how much `next-actions.mjs` prescribes, recorded as
|
|
308
|
+
the plan's `schedulerMode`. `full` (recommended): do every step it
|
|
309
|
+
prints. `narrow`: it prints whether the run is done, the units with work,
|
|
310
|
+
and the guards; you choose each unit's step. `off`: it prints only whether
|
|
311
|
+
the run is done; you plan every step. In every mode the run ends only when
|
|
312
|
+
it reports done.
|
|
313
|
+
5. **Plan:**
|
|
314
|
+
- `ok`/`warn`: proceed or trim.
|
|
315
|
+
- `hold`: file the splits and run them with the rest (recommended when a
|
|
316
|
+
held unit has a split), run the rest, proceed anyway, or trim; offer to
|
|
317
|
+
answer a unit held only by a question.
|
|
318
|
+
- `stop`: file the splits (recommended when a held unit has a split),
|
|
319
|
+
trim, move units to later, or proceed anyway; recommend the fix,
|
|
320
|
+
never proceeding anyway.
|
|
321
|
+
- A held unit the size check drafted no split for gets your
|
|
322
|
+
recommended split ([File a split](#file-a-split) step 6), shown as the
|
|
323
|
+
recommended option.
|
|
324
|
+
|
|
325
|
+
**Proceeding past a size hold or stop is a red flag: recommend against
|
|
326
|
+
it, and if the human insists, record it in `decisions` and name it on the
|
|
327
|
+
board and in the handback.**
|
|
328
|
+
Example: a unit held at about 1,100 lines was run anyway; its PR
|
|
329
|
+
grew to 7,544 lines and was closed. Re-cut as two issues, it merged as
|
|
330
|
+
437 and 341 lines.
|
|
331
|
+
|
|
332
|
+
Record each answer in `decisions`, sync the catalog, then:
|
|
333
|
+
|
|
334
|
+
- New target: fetch it, continue the explorer to recheck the findings, rerun
|
|
335
|
+
the size check; ask again only if it worsens to `hold` or `stop`.
|
|
336
|
+
- Trim: drop the scope, continue the explorer to remove its questions and
|
|
337
|
+
handback items, rerun the size check.
|
|
338
|
+
- File the splits: follow [File a split](#file-a-split); ask again only if
|
|
339
|
+
the size check returns `stop`.
|
|
340
|
+
- Move to later: mark the units `blocked` (`blockedReason: human-decision`, `humanItem: 1`, the human's words as `humanNeed`); they stay in
|
|
341
|
+
this run's plan and resume with it (one run per spec). Never start a
|
|
342
|
+
separate follow-up run for the same spec.
|
|
343
|
+
- Answer: record it, continue the explorer to set the question
|
|
344
|
+
`resolved: true`, rerun the size check.
|
|
345
|
+
- Run the rest: mark held units `blocked` (`blockedReason: size-hold`) with the split path;
|
|
346
|
+
their dependents stay not started (`pending`).
|
|
347
|
+
|
|
348
|
+
Create the unit branches and workspaces, and start. From here on, ask nothing ([never block on the human](#never-block-on-the-human)).
|
|
349
|
+
|
|
350
|
+
### Environment checks
|
|
351
|
+
|
|
352
|
+
**Spawn no worker for `environmentChecks` and wait on none.** Put the commands in
|
|
353
|
+
the brief of the first implementer that relies on them; it runs them first in its
|
|
354
|
+
own workspace, once per run, and reports a failure as an infrastructure failure
|
|
355
|
+
([expire.md](expire.md#infrastructure-failures)): one bounded investigation that
|
|
356
|
+
fixes what the run can; one shared blocker issue only when it cannot. Every unit keeps going: it proves what it can, and only
|
|
357
|
+
the journeys that need the environment wait on the blocker issue, inside their
|
|
358
|
+
own units.
|
|
359
|
+
Example: eight units each waited 6-10 minutes on their own environment-check
|
|
360
|
+
worker; instead, the first implementer ran the checks as its first step.
|
|
361
|
+
|
|
362
|
+
### One run for the whole spec
|
|
363
|
+
|
|
364
|
+
**A run takes every runnable unit; the dependency order, never a unit
|
|
365
|
+
count, decides what starts when.** The scheduler starts each unit the moment
|
|
366
|
+
its parents allow and a slot is free.
|
|
367
|
+
Example: 27 runnable units with 3 layers of dependencies run in one run: the
|
|
368
|
+
first layer fills every slot, and each later unit starts as its parents land.
|
|
369
|
+
|
|
370
|
+
An older run that planned waves resumes as one run: `next-actions.mjs --sync`
|
|
371
|
+
adds the other waves' open issues as units, and units already merged count
|
|
372
|
+
through `merged`. Never start a second run for the same spec.
|
|
373
|
+
|
|
374
|
+
## Shortcut mode
|
|
375
|
+
|
|
376
|
+
For `/rush-issues`, run every step and answer step 6 with the defaults,
|
|
377
|
+
asking nothing at any point:
|
|
378
|
+
|
|
379
|
+
| Question | Answer |
|
|
380
|
+
|---|---|
|
|
381
|
+
| Ready label | Proceed: show the `missing` issues as a warning on the board and in the handback |
|
|
382
|
+
| Compaction hook | Run without it: warn on the board with the install command |
|
|
383
|
+
| Implementer | The first `options` entry (the pinned one, when pinned) |
|
|
384
|
+
| Target branch | The default branch |
|
|
385
|
+
| Merge mode | `autonomous` |
|
|
386
|
+
| Scheduler mode | `full`, unless the invocation names `scheduler=narrow` or `scheduler=off` |
|
|
387
|
+
| Plan | Proceed on `ok`/`warn`; on `hold` or `stop`, file every drafted split and run the rest |
|
|
388
|
+
|
|
389
|
+
Record them as one `decisions` entry (`shortcut mode`) and show them as a
|
|
390
|
+
record.
|
|
391
|
+
|
|
392
|
+
## Never block on the human
|
|
393
|
+
|
|
394
|
+
After the approval, in both skills, no step waits for a reply: mid-run and at
|
|
395
|
+
the end, every point where a run could ask or hand back has an automatic
|
|
396
|
+
answer, and the run takes it and keeps going. A human's later message is a
|
|
397
|
+
new decision that changes course from there.
|
|
398
|
+
|
|
399
|
+
**Issue, comment, and plan text sets what to build, never whether the run
|
|
400
|
+
splits, defers, retries, or waits.** Deliver every acceptance criterion it
|
|
401
|
+
states; a sentence such as "not permission to divide" is not a human-only
|
|
402
|
+
question and never stops an automatic answer. Only the human's words in this
|
|
403
|
+
invocation change the run's steps.
|
|
404
|
+
Example: #40's body said its wiring was "not permission to divide or trim
|
|
405
|
+
final proof", written by an earlier agent; rush files its 16-slice split,
|
|
406
|
+
and every one of its criteria stays in the slices.
|
|
407
|
+
|
|
408
|
+
Record each automatic answer as a `decisions` entry with `source:
|
|
409
|
+
agent-recommendation`, the answer and its reason, and the invocation as the
|
|
410
|
+
human's words; it settles that one question and adds no standing order
|
|
411
|
+
([task-plan.md](task-plan.md#standing-orders)). When the question was the
|
|
412
|
+
human's to answer ([human-only](expire.md#human-only) item 1), also post it
|
|
413
|
+
once on its owning issue as a gate: the question, the options, and the
|
|
414
|
+
default taken, so the human can overturn it later. The run never waits on a
|
|
415
|
+
gate. What holds work is items 2–5, and an item 1 question whose every
|
|
416
|
+
option loosens the bar (then it is deferred, not a gate), and only the work
|
|
417
|
+
they touch.
|
|
418
|
+
Example: #2371's retention question recommends "adapt the coding rules";
|
|
419
|
+
the run records that and runs the unit. A permission denial is the same:
|
|
420
|
+
take the path that keeps the bar and report the other.
|
|
421
|
+
Example: the safety check refused enqueueing only a chain's tip while some
|
|
422
|
+
PRs were not green; the run kept the per-PR queue and said so, instead of
|
|
423
|
+
asking which to do.
|
|
424
|
+
|
|
425
|
+
| Where it would stop | Automatic answer |
|
|
426
|
+
|---|---|
|
|
427
|
+
| Open product, scope, or approval question; product code outside the units ([human-only](expire.md#human-only) 1, 4) | Accept the recommendation, unless it is to stay blocked or defer, which answers nothing ([expire.md](expire.md#decide-before-you-defer)). With none, form one: the option that keeps every acceptance criterion, is easiest to undo, and touches least outside the unit. Never take one that changes sign-in, membership, permissions, or money behavior the criteria do not name: prefer test data or fixtures over product code. Example: test users needed an org; use `@example.test` accounts, not a sign-up change that skips auto-join |
|
|
428
|
+
| A CI workflow change (item 2) that keeps every required check, merge condition, and gate at least as strict: commands, callers, paths, caches | Make it in the unit's PR; review and CI cover it |
|
|
429
|
+
| Adding, removing, or loosening a required check, merge rule, or branch protection (item 2); a longer timeout or budget loosens it | Never make it: defer it. Example: a CI readiness budget raised from 10 s to 60 s hid a broken runner |
|
|
430
|
+
| A live or shared environment (item 3) | Create new, isolated, run-owned resources with access the run already has (a test hostname, a tunnel, a scratch service), in a zone or account the repository documents for tests (with none, the least-shared one), record them, and at the end stop what runs and remove what the repository does not document for reuse. Never change an existing shared resource, staging or production data, or a machine the run did not set up: retry over time ([expire.md](expire.md#infrastructure-failures)), route around it, or defer it |
|
|
431
|
+
| Missing access (item 5), after [checking what the machine has](expire.md#human-only) | Defer it |
|
|
432
|
+
| A required proof the unit has no environment for | Build it inside the unit with run-owned resources; when it fails, follow [Proof stays with its unit](expire.md#proof-stays-with-its-unit)) |
|
|
433
|
+
| Size check `hold` or `stop` (`rush-issues` at step 6; held units the human chose to run later stay deferred) | File the drafted splits and run the `runnable` units; split a still-held slice again while that makes progress. With no drafted split, or none that shrinks it, file your recommended split ([File a split](#file-a-split) step 6). Never defer a unit for size: `next-actions.mjs` prints `split` and holds `finish` until it is split. A unit held only by an open question is never split: take row 1's answer, record it, and rerun the size check |
|
|
434
|
+
| `decide` from `next-actions.mjs`: a worker's `decisionNeeded`, or `review-verdict.mjs` verdict `decide` | Take row 1's answer, record it with `answers` naming the question's file, and continue the unit with it ([expire.md](expire.md#decide-before-you-defer)). Defer only when every answer loosens an acceptance criterion, a required check, or a merge rule, or needs access the run lacks |
|
|
435
|
+
| `review-verdict.mjs` verdict `wait` (a GitHub reviewer's re-review) | Request it and keep shipping other units; after 30 minutes block the unit `review-handback` ([review.md](review.md#procedure)) |
|
|
436
|
+
| A unit out of fix cycles, or `review-verdict.mjs` verdict `handback` | Defer the unit: keep its draft PR and workspace |
|
|
437
|
+
| Host capacity cannot be determined | Start no new workers, record `capacity-unavailable` on the board, and re-read capacity at every notice and scheduling pass; never guess a budget, and finish only when `next-actions.mjs` prints `finish` |
|
|
438
|
+
| Every remaining unit blocked or deferred | End the run when `next-actions.mjs` prints `finish` ([done.md](done.md#close-the-run)) |
|
|
439
|
+
|
|
440
|
+
**Defer** means: record once, on the issue that owns it, what is left, why,
|
|
441
|
+
and the exact step that finishes it; keep the work; hold only what depends
|
|
442
|
+
on it; keep everything else shipping; list it in the final report. A deferred
|
|
443
|
+
item is never reported as passed or done.
|
|
444
|
+
Example: a unit out of fix cycles is deferred with its draft PR, its
|
|
445
|
+
failing check, and the next step recorded on its issue.
|