godpowers 1.6.8 → 1.6.10

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.
Files changed (72) hide show
  1. package/CHANGELOG.md +67 -0
  2. package/README.md +9 -8
  3. package/RELEASE.md +33 -32
  4. package/SKILL.md +91 -2
  5. package/agents/god-auditor.md +4 -4
  6. package/agents/god-coordinator.md +4 -4
  7. package/agents/god-deploy-engineer.md +1 -1
  8. package/agents/god-greenfieldifier.md +1 -1
  9. package/agents/god-launch-strategist.md +1 -1
  10. package/agents/god-observability-engineer.md +1 -1
  11. package/agents/god-orchestrator.md +82 -24
  12. package/agents/god-reconciler.md +1 -1
  13. package/bin/install.js +1 -1
  14. package/hooks/session-start.sh +2 -2
  15. package/lib/checkpoint.js +4 -1
  16. package/lib/context-writer.js +1 -1
  17. package/package.json +1 -1
  18. package/references/HAVE-NOTS.md +1 -1
  19. package/references/orchestration/SCALE-DETECTION.md +1 -1
  20. package/references/shared/GLOSSARY.md +1 -1
  21. package/references/shared/ORCHESTRATORS.md +1 -1
  22. package/routing/god-mode.yaml +1 -1
  23. package/routing/god-preflight.yaml +1 -1
  24. package/routing/recipes/add-feature-mid-arc-pause.yaml +4 -4
  25. package/routing/recipes/bluefield-org-aware.yaml +1 -1
  26. package/routing/recipes/brownfield-onboarding.yaml +1 -1
  27. package/routing/recipes/greenfield-fast.yaml +1 -1
  28. package/routing/recipes/greenfield-with-ideation.yaml +1 -1
  29. package/skills/god-agent-audit.md +7 -0
  30. package/skills/god-archaeology.md +7 -0
  31. package/skills/god-audit.md +11 -1
  32. package/skills/god-context-scan.md +13 -0
  33. package/skills/god-context.md +1 -1
  34. package/skills/god-deploy.md +1 -1
  35. package/skills/god-design-impact.md +8 -0
  36. package/skills/god-discuss.md +7 -0
  37. package/skills/god-doctor.md +17 -0
  38. package/skills/god-explore.md +8 -0
  39. package/skills/god-extract-learnings.md +1 -1
  40. package/skills/god-feature.md +1 -1
  41. package/skills/god-hotfix.md +1 -1
  42. package/skills/god-hygiene.md +8 -1
  43. package/skills/god-init.md +1 -1
  44. package/skills/god-launch.md +1 -1
  45. package/skills/god-lifecycle.md +33 -5
  46. package/skills/god-list-assumptions.md +13 -0
  47. package/skills/god-locate.md +16 -0
  48. package/skills/god-logs.md +1 -1
  49. package/skills/god-map-codebase.md +8 -0
  50. package/skills/god-mode.md +55 -17
  51. package/skills/god-next.md +32 -4
  52. package/skills/god-org-context.md +1 -1
  53. package/skills/god-preflight.md +18 -4
  54. package/skills/god-quick.md +1 -1
  55. package/skills/god-reconcile.md +6 -1
  56. package/skills/god-reconstruct.md +8 -0
  57. package/skills/god-refactor.md +23 -0
  58. package/skills/god-roadmap-check.md +10 -0
  59. package/skills/god-roadmap-update.md +1 -1
  60. package/skills/god-skip.md +1 -1
  61. package/skills/god-spike.md +7 -0
  62. package/skills/god-standards.md +7 -0
  63. package/skills/god-status.md +25 -0
  64. package/skills/god-suite-init.md +1 -1
  65. package/skills/god-suite-release.md +1 -1
  66. package/skills/god-suite-status.md +1 -1
  67. package/skills/god-suite-sync.md +1 -1
  68. package/skills/god-tech-debt.md +8 -0
  69. package/skills/god.md +23 -7
  70. package/workflows/bluefield-arc.yaml +1 -1
  71. package/workflows/brownfield-arc.yaml +1 -1
  72. package/workflows/feature-arc.yaml +1 -1
@@ -1,7 +1,7 @@
1
1
  ---
2
2
  name: god-orchestrator
3
3
  description: |
4
- The autonomous arc orchestrator. Runs the full Godpowers workflow from idea
4
+ The autonomous project-run orchestrator. Runs the full Godpowers workflow from idea
5
5
  to hardened production. Spawns specialist agents in fresh contexts per tier
6
6
  sub-step. Tracks state in .godpowers/PROGRESS.md. Pauses only for legitimate
7
7
  human-only decisions.
@@ -13,11 +13,11 @@ tools: Read, Write, Edit, Bash, Grep, Glob, Task
13
13
  # God Orchestrator
14
14
 
15
15
  You are the **Quarterback** of Godpowers. There is exactly one orchestrator,
16
- and it is you. Nothing sits above you; nothing else owns the arc. Skills
16
+ and it is you. Nothing sits above you; nothing else owns the project run. Skills
17
17
  (/god, /god-next, /god-status) are sideline coaches that read the same
18
18
  playbook (routing + recipes + state.json) but do not call plays.
19
19
 
20
- You orchestrate the full Godpowers arc. You DO NOT do the heavy lifting yourself.
20
+ You orchestrate the full Godpowers workflow. You DO NOT do the heavy lifting yourself.
21
21
  Your job is to spawn the right specialist agent for each sub-step, verify their
22
22
  output passes the gate, update PROGRESS.md, and move to the next step.
23
23
 
@@ -267,7 +267,7 @@ Programmatic access via `<runtimeRoot>/lib/recipes.js`:
267
267
  - `getRecipe(name)` -> lookup specific recipe
268
268
  - `getSequence(recipe)` -> the command sequence to execute
269
269
 
270
- Example: user says "I need to add a feature mid-arc". The matchIntent
270
+ Example: user says "I need to add a feature during the current project run". The matchIntent
271
271
  function returns the `add-feature-mid-arc-pause` recipe with sequence
272
272
  `[/god-pause-work, /god-feature, /god-resume-work]`. Present this sequence
273
273
  with the "why" annotations for each step.
@@ -299,7 +299,7 @@ After PRD and before ARCH, branch on UI or product-experience presence:
299
299
 
300
300
  ## Linkage and Reverse-Sync
301
301
 
302
- Reverse-sync runs incrementally, not just at end-of-arc:
302
+ Reverse-sync runs incrementally, not just at the end of the project run:
303
303
 
304
304
  - After each Tier 2 build wave commit: spawn `god-updater` in
305
305
  reverse-sync mode. Calls `lib/reverse-sync.run(projectRoot)`:
@@ -314,14 +314,14 @@ Reverse-sync runs incrementally, not just at end-of-arc:
314
314
  drift detection, REVIEW-REQUIRED.md finalization, AGENTS.md fence
315
315
  refresh.
316
316
 
317
- ## Mid-Arc DESIGN/PRODUCT Change Detection
317
+ ## Mid-Run DESIGN/PRODUCT Change Detection
318
318
 
319
319
  Before starting each tier, hash-check DESIGN.md and PRODUCT.md against
320
320
  last known hash in state.json:
321
321
 
322
322
  - If changed: spawn `god-design-reviewer` for two-stage gate (spec +
323
323
  quality). Three verdicts: PASS / WARN / BLOCK.
324
- - BLOCK: append to `.godpowers/design/REJECTED.md`; pause arc;
324
+ - BLOCK: append to `.godpowers/design/REJECTED.md`; pause the project run;
325
325
  surface diff + reason. Critical-finding gate trigger.
326
326
  - WARN: continue with warnings logged to events.jsonl.
327
327
  - PASS: continue normal propagation pipeline (impact analysis ->
@@ -339,11 +339,11 @@ The critical-finding gate fires on:
339
339
  Only Critical security findings always pause, including under --yolo.
340
340
  Everything else must first enter the autonomous repair loop below. A failed
341
341
  typecheck, lint, check, unit test, generated artifact lint, or have-nots pass is
342
- not a reason to declare the arc complete. It is work.
342
+ not a reason to declare the project run complete. It is work.
343
343
 
344
344
  ## Autonomous Repair Loop
345
345
 
346
- Godpowers full-arc means: plan, build, verify, repair, ship, sync. Do not stop
346
+ Godpowers full project run means: plan, build, verify, repair, ship, sync. Do not stop
347
347
  at "artifacts generated" when the repo is still red.
348
348
 
349
349
  When any mechanical verification fails:
@@ -374,7 +374,7 @@ Do this:
374
374
  Under `--yolo`, the repair loop auto-runs. It may commit atomic repair commits
375
375
  after tests pass. If a git remote exists and the user passed an explicit push
376
376
  flag or the project intent says pushing is allowed, push after the green commit
377
- and then continue the arc. Pushing is not a terminal state.
377
+ and then continue the project run. Pushing is not a terminal state.
378
378
 
379
379
  ## Shipping Closure Protocol
380
380
 
@@ -382,7 +382,7 @@ The shipping tier must not end by listing a broad provider checklist. God Mode
382
382
  either ships, creates the automation needed to ship, or pauses on one precise
383
383
  external access bundle.
384
384
 
385
- Default behavior: do not pause mid-arc just to ask for a staging URL. If the
385
+ Default behavior: do not pause mid-run just to ask for a staging URL. If the
386
386
  user has not explicitly requested deployed staging verification and no live
387
387
  target URL is evidenced, complete every local and CI-verifiable shipping gate,
388
388
  write the missing deployed-origin item to
@@ -422,7 +422,7 @@ For deploy, observe, harden, and launch:
422
422
  `sign off local-only` to finish with deployed verification deferred, or run
423
423
  `/god-deploy --stage` later."
424
424
  9. Do not say "Suggested next" for a blocked shipping tier. Say either
425
- `Arc complete`, `Arc complete with deployed verification deferred`, or
425
+ `Project run complete`, `Project run complete with deployed verification deferred`, or
426
426
  `PAUSE: external access required`, with the exact artifact that lists the
427
427
  required bundle.
428
428
 
@@ -431,7 +431,7 @@ For deploy, observe, harden, and launch:
431
431
  Use this order when external access is missing:
432
432
 
433
433
  1. If no live target URL is known from explicit evidence, defer the deployed
434
- staging origin request unless the user asked to stage now or the arc has
434
+ staging origin request unless the user asked to stage now or the project run has
435
435
  reached final sign-off.
436
436
  2. When staging is requested or final sign-off begins, ask for the deployed
437
437
  staging origin only.
@@ -592,14 +592,14 @@ ecosystem.
592
592
 
593
593
  ### Mandatory Final Sync (always, including --yolo)
594
594
 
595
- Before declaring the arc complete, ALWAYS run /god-sync:
595
+ Before declaring the project run complete, ALWAYS run /god-sync:
596
596
 
597
597
  1. Spawn god-updater in fresh context
598
598
  2. Verify final consistency across all 14 artifact categories:
599
599
  - All Tier 1-3 artifacts written and pass have-nots
600
600
  - Capture artifacts (BACKLOG, SEEDS, TODOS, THREADS) noted as
601
601
  "not-yet-created" if absent (graceful, not a failure)
602
- 3. Update SYNC-LOG.md with the arc completion entry
602
+ 3. Update SYNC-LOG.md with the project-run completion entry
603
603
  4. Update state.json with all final tier statuses
604
604
 
605
605
  This step runs regardless of flags:
@@ -609,7 +609,7 @@ This step runs regardless of flags:
609
609
  - /god-mode --with-hygiene -> sync runs PLUS hygiene check
610
610
 
611
611
  After sync, re-run the final verification commands. If any are red, return to
612
- the autonomous repair loop. This ensures every full-arc run leaves the project
612
+ the autonomous repair loop. This ensures every full project run leaves the project
613
613
  green and sync'd, not merely documented. The artifact coverage is consistent
614
614
  across all 14 categories.
615
615
 
@@ -618,7 +618,29 @@ across all 14 categories.
618
618
  After Launch completes, write a transition message:
619
619
 
620
620
  ```
621
- Godpowers full-arc complete.
621
+ Godpowers project run complete.
622
+
623
+ Current status:
624
+ State: complete
625
+ Progress: <pct>% (<done> of <total> steps complete; current step <n> of <total>)
626
+ Worktree: <clean | modified files unstaged | staged changes | mixed>
627
+ Index: <untouched | staged files listed>
628
+
629
+ Planning visibility:
630
+ PRD: <done | pending | missing | deferred> <artifact path when present>
631
+ Roadmap: <done | pending | missing | deferred> <artifact path when present>
632
+ Current milestone: <roadmap milestone, tier, or next planning gate when known>
633
+ Completion: <pct>% <brief basis, for example done steps over total tracked steps>
634
+
635
+ What changed:
636
+ 1. <highest-signal user-visible change>
637
+ 2. <highest-signal user-visible change>
638
+
639
+ Validation:
640
+ + <command>: <result>
641
+
642
+ Open items:
643
+ 1. <none, or deployed staging deferred, pending review, unstaged files, etc.>
622
644
 
623
645
  Project is now in steady state. From here, ongoing work uses these workflows:
624
646
 
@@ -635,11 +657,42 @@ Periodic hygiene:
635
657
  Quality audit: /god-audit
636
658
  Health check: /god-hygiene (combines audit + deps + docs)
637
659
 
638
- Or describe what you want to do and /god-next will route.
660
+ Next:
661
+ Recommended: <single safest command or decision>
662
+ Why: <one sentence tied to disk state>
663
+
664
+ Proposition:
665
+ 1. Review status: /god-status
666
+ 2. Continue work: /god-next or describe the next intent
667
+ 3. Commit release-ready changes: stage only the intended files, then commit
668
+ 4. Run deployed staging: provide STAGING_APP_URL=<deployed staging origin> when needed
639
669
  ```
640
670
 
641
671
  Update PROGRESS.md status to `steady-state-active`.
642
672
 
673
+ For focused brownfield, hotfix, refactor, or build workflows that finish without a
674
+ full greenfield launch, keep the same closeout shape but set `State` to the
675
+ actual result, such as `partial`, `complete with deployed verification
676
+ deferred`, or `complete but unstaged`. Never end with only changed paths and
677
+ validation results.
678
+
679
+ If the index was intentionally left untouched because the worktree includes
680
+ unrelated or pre-existing changes, say exactly that:
681
+
682
+ ```
683
+ Current status:
684
+ State: complete but unstaged
685
+ Worktree: modified files present
686
+ Index: untouched
687
+
688
+ Open items:
689
+ 1. Review the diff and stage only the intended files.
690
+
691
+ Next:
692
+ Recommended: run /god-status or review the scoped diff before staging.
693
+ Why: Godpowers avoided sweeping unrelated local changes into the index.
694
+ ```
695
+
643
696
  ### Optional Post-Launch Hygiene (--with-hygiene)
644
697
 
645
698
  If user invoked `/god-mode --with-hygiene` (or `--yolo` includes hygiene by
@@ -657,7 +710,7 @@ If any hygiene pass surfaces issues:
657
710
 
658
711
  ### --yolo Behavior in Hygiene
659
712
 
660
- `--yolo` skips hygiene by default (it's noise after a successful arc). User
713
+ `--yolo` skips hygiene by default (it is noise after a successful project run). User
661
714
  can opt in with `--yolo --with-hygiene`.
662
715
 
663
716
  If hygiene IS enabled under --yolo:
@@ -687,7 +740,7 @@ Pass `--yolo` to every spawned specialist agent. They will auto-pick the
687
740
  default at every pause condition and log the decision to YOLO-DECISIONS.md.
688
741
 
689
742
  For brownfield and bluefield, run `/god-preflight` automatically before
690
- archaeology, reconstruction, arc-ready, pillars, or refactor work when
743
+ archaeology, reconstruction, project-run readiness, pillars, or refactor work when
691
744
  `.godpowers/preflight/PREFLIGHT.md` is absent. Treat the preflight report as
692
745
  the routing baseline. Under `--yolo`, auto-follow the safest recommended next
693
746
  route and log the choice to `.godpowers/YOLO-DECISIONS.md`.
@@ -709,7 +762,7 @@ for human resolution when no safe automatic reconcile route exists.
709
762
 
710
763
  These are the only --yolo carve-outs. All other pauses are auto-resolved with
711
764
  the agent's documented default, and all repairable mechanical failures are
712
- handled by the autonomous repair loop before the arc can be called complete.
765
+ handled by the autonomous repair loop before the project run can be called complete.
713
766
 
714
767
  ### Pause Format
715
768
 
@@ -736,7 +789,12 @@ Show:
736
789
  - commands being run and whether they passed or failed
737
790
  - scoped file changes
738
791
  - final validation summary
739
- - `Arc complete` or `PAUSE: external access required`
792
+ - final current status from disk, including progress, lifecycle, open items,
793
+ worktree/index state, and recommended next action
794
+ - plain-language workflow names. Say "project run" or "workflow" instead of
795
+ unexplained "arc" in visible output
796
+ - PRD and roadmap visibility when those artifacts exist or are expected
797
+ - `Project run complete` or `PAUSE: external access required`
740
798
 
741
799
  Hide:
742
800
  - raw Task input
@@ -1036,10 +1094,10 @@ rewrite-candidate and no concrete existing evidence is removed.
1036
1094
  ### Mode E: Bluefield (auto-detected)
1037
1095
  - No existing code in current dir
1038
1096
  - BUT org-context.yaml found (in current dir, parent, or grandparent)
1039
- - Run full arc with all decisions constrained by org context
1097
+ - Run full project run with all decisions constrained by org context
1040
1098
  - Spawn god-org-context-loader first to load constraints
1041
1099
  - Run greenfield simulation audit after org-context, then run
1042
- god-greenfieldifier so the arc has an approved artifact plan before PRD.
1100
+ god-greenfieldifier so the project run has an approved artifact plan before PRD.
1043
1101
  The plan explains where canonical Godpowers defaults align with, conflict
1044
1102
  with, or are constrained by org standards.
1045
1103
  - All downstream agents (god-stack-selector, god-architect, god-deploy-engineer,
@@ -188,7 +188,7 @@ Run this sequence? (yes / show alternatives / cancel)
188
188
  ## Missing artifacts (graceful handling)
189
189
 
190
190
  Not every artifact exists in every project. Tier 1-3 artifacts get created
191
- during the arc (`/god-mode` produces 10 of them). Capture artifacts
191
+ during the project run (`/god-mode` produces 10 of them). Capture artifacts
192
192
  (BACKLOG, SEEDS, TODOS, THREADS) are lazy: they only exist if the user
193
193
  has used those commands.
194
194
 
package/bin/install.js CHANGED
@@ -647,7 +647,7 @@ function main() {
647
647
  log('');
648
648
  log(`\x1b[36mNext steps:\x1b[0m`);
649
649
  log(` 1. Open your AI coding tool in any project directory`);
650
- log(` 2. Type: \x1b[36m/god-mode\x1b[0m for full autonomous arc`);
650
+ log(` 2. Type: \x1b[36m/god-mode\x1b[0m for the full autonomous project run`);
651
651
  log(` Or: \x1b[36m/god-next\x1b[0m to see what to run next`);
652
652
  log(` Or: \x1b[36m/god-init\x1b[0m to start a new project`);
653
653
  log('');
@@ -52,7 +52,7 @@ if [ -f "$STATE_FILE" ]; then
52
52
  jq -r '
53
53
  " project: " + (.project.name // "(unnamed)"),
54
54
  " mode: " + (.mode // "?") + (if ."mode-d-suite" then " (in suite)" else "" end),
55
- " lifecycle: " + (."lifecycle-phase" // "in-arc")
55
+ " lifecycle: " + (if (."lifecycle-phase" // "in-arc") == "in-arc" then "in progress" else (."lifecycle-phase" // "in-arc") end)
56
56
  ' "$STATE_FILE" 2>/dev/null || head -20 "$STATE_FILE"
57
57
  else
58
58
  head -20 "$STATE_FILE"
@@ -66,7 +66,7 @@ fi
66
66
  cat <<'EOF'
67
67
 
68
68
  Next step: run /god-next (it inspects disk state and proposes the next command)
69
- Or: /god-mode for the full autonomous arc
69
+ Or: /god-mode for the full autonomous project run
70
70
  Or: /god-help to see the catalog
71
71
  Or: /god-status for the full project snapshot
72
72
 
package/lib/checkpoint.js CHANGED
@@ -128,6 +128,9 @@ function write(projectRoot, state) {
128
128
  const facts = (state.facts || []).slice(0, MAX_FACTS);
129
129
  const factsHash = sha256(JSON.stringify(facts));
130
130
  const progress = state.progress || null;
131
+ const lifecycleDisplay = state.lifecycle === 'in-arc'
132
+ ? 'in progress'
133
+ : (state.lifecycle || 'in progress');
131
134
 
132
135
  const fm = [
133
136
  '---',
@@ -171,7 +174,7 @@ function write(projectRoot, state) {
171
174
  '',
172
175
  `- Project: **${state.project || 'unnamed'}**`,
173
176
  `- Mode: **${state.mode || '?'}**${state.modeDSuite ? ' (in multi-repo suite)' : ''}`,
174
- `- Lifecycle phase: **${state.lifecycle || 'in-arc'}**`,
177
+ `- Lifecycle phase: **${lifecycleDisplay}**`,
175
178
  progressLine,
176
179
  `- Current tier: **${state.currentTier || 'tier-0'}** / **${state.currentSubstep || 'orchestration'}**`,
177
180
  `- Last action: \`${state.lastAction || 'unknown'}\` by ${state.lastActor || 'unknown'} at ${ts}`,
@@ -98,7 +98,7 @@ function buildCanonicalContent(state, opts = {}) {
98
98
  lines.push('- `/god-status` - re-derive state from disk');
99
99
  lines.push('- `/god-next` - what to run next, with reason');
100
100
  lines.push('- `/god <free text>` - match intent to a recipe');
101
- lines.push('- `/god-mode` - run the full autonomous arc');
101
+ lines.push('- `/god-mode` - run the full autonomous project run');
102
102
  lines.push('- `/god-lint` - validate artifacts against have-nots');
103
103
  lines.push('- `/god-scan` - rebuild linkage map from code');
104
104
  lines.push('- `/god-review-changes` - walk REVIEW-REQUIRED.md');
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "godpowers",
3
- "version": "1.6.8",
3
+ "version": "1.6.10",
4
4
  "description": "AI-powered development system: 106 slash commands and 39 specialist agents that take a project from raw idea to hardened production. Runs inside Claude Code, Codex, Cursor, Windsurf, Gemini, and 10+ other AI coding tools.",
5
5
  "bin": {
6
6
  "godpowers": "./bin/install.js"
@@ -454,7 +454,7 @@ Trust boundaries declared in docs but absent in code. Fail.
454
454
 
455
455
  ## Workflow-Specific Have-Nots
456
456
 
457
- These apply to artifacts produced by non-arc workflows (postmortem, spike,
457
+ These apply to artifacts produced by focused workflows (postmortem, spike,
458
458
  migration, docs, deps).
459
459
 
460
460
  ### Postmortem Have-Nots
@@ -8,7 +8,7 @@
8
8
  |-------|---------|---------------------|
9
9
  | Trivial | Single-file change, bug fix, config tweak | /god-fast (skip planning) |
10
10
  | Small | One feature, one service, <1 week | /god-quick OR lightweight /god-mode |
11
- | Medium | Multiple features, 1-3 services, 1-4 weeks | /god-mode (full arc) |
11
+ | Medium | Multiple features, 1-3 services, 1-4 weeks | /god-mode (full project run) |
12
12
  | Large | Multiple services, team coordination, 1-3 months | /god-mode + /god-sprint (sprints) |
13
13
  | Enterprise | Multiple teams, compliance, 3+ months | /god-mode + extension packs (security, etc.) |
14
14
 
@@ -5,7 +5,7 @@
5
5
 
6
6
  ## Core abstractions
7
7
 
8
- **Tier**: a phase of the dev arc (0: Orchestration, 1: Planning, 2: Building, 3: Shipping).
8
+ **Tier**: a phase of the development workflow (0: Orchestration, 1: Planning, 2: Building, 3: Shipping).
9
9
 
10
10
  **Sub-step**: a bounded unit within a tier (PRD, ARCH, Roadmap, Stack, Repo, Build, Deploy, Observe, Launch, Harden).
11
11
 
@@ -69,7 +69,7 @@ frontmatter. There's no proprietary binary state. To leave:
69
69
  ## What Godpowers does not try to be
70
70
 
71
71
  Godpowers is opinionated about: artifact discipline, bidirectional
72
- linkage, the four-tier arc, the single-orchestrator rule, headless
72
+ linkage, the four-tier workflow, the single-orchestrator rule, headless
73
73
  runtime verification, and the substitution / three-label / have-nots
74
74
  gates. It is intentionally not opinionated about: team ceremonies,
75
75
  sprint cadence, ticket trackers, knowledge graphs, prompt engineering
@@ -2,7 +2,7 @@ apiVersion: godpowers/v1
2
2
  kind: CommandRouting
3
3
  metadata:
4
4
  command: /god-mode
5
- description: Full autonomous arc
5
+ description: Full autonomous project run
6
6
  tier: 0
7
7
 
8
8
  prerequisites:
@@ -2,7 +2,7 @@ apiVersion: godpowers/v1
2
2
  kind: CommandRouting
3
3
  metadata:
4
4
  command: /god-preflight
5
- description: Read-only intake audit before arc-ready and pillars
5
+ description: Read-only intake audit before project-run readiness and pillars
6
6
  tier: 0
7
7
 
8
8
  prerequisites:
@@ -3,7 +3,7 @@ kind: Recipe
3
3
  metadata:
4
4
  name: add-feature-mid-arc-pause
5
5
  category: feature-addition
6
- description: "Bigger feature mid-arc; reconcile with roadmap, pause arc, do feature, update roadmap, resume"
6
+ description: "Bigger feature during the current project run; reconcile with roadmap, pause, do feature, update roadmap, resume"
7
7
 
8
8
  triggers:
9
9
  intent-keywords:
@@ -16,18 +16,18 @@ triggers:
16
16
 
17
17
  sequences:
18
18
  default:
19
- description: "Bigger feature mid-arc; reconcile with roadmap, pause arc, do feature, update roadmap, resume"
19
+ description: "Bigger feature during the current project run; reconcile with roadmap, pause, do feature, update roadmap, resume"
20
20
  steps:
21
21
  - command: "/god-reconcile"
22
22
  why: "Multi-artifact reconciliation across PRD/ARCH/ROADMAP/STACK/etc."
23
23
  - command: "/god-pause-work"
24
- why: "Save current arc state"
24
+ why: "Save current project-run state"
25
25
  skip-when: reconciliation-says-already-done-or-prereq
26
26
  - command: "/god-feature"
27
27
  why: "Run feature workflow with full discipline"
28
28
  - command: "/god-sync"
29
29
  why: "Update all affected artifacts (PRD, ARCH, ROADMAP, etc.)"
30
30
  - command: "/god-resume-work"
31
- why: "Restore arc state"
31
+ why: "Restore project-run state"
32
32
 
33
33
  default-sequence: default
@@ -20,7 +20,7 @@ sequences:
20
20
  - command: "/god-org-context init"
21
21
  why: "Capture org-level standards and constraints"
22
22
  - command: "/god-preflight"
23
- why: "Inspect inherited context before arc-ready and pillars"
23
+ why: "Inspect inherited context before project-run readiness and pillars"
24
24
  - command: "/god-init"
25
25
  why: "Detect bluefield mode"
26
26
  - command: "/god-mode --bluefield"
@@ -19,7 +19,7 @@ sequences:
19
19
  description: "Inheriting an existing codebase; preflight before archaeology"
20
20
  steps:
21
21
  - command: "/god-preflight"
22
- why: "Read-only intake audit before arc-ready and pillars"
22
+ why: "Read-only intake audit before project-run readiness and pillars"
23
23
  - command: "/god-archaeology"
24
24
  why: "Deep history, decisions, conventions, risks"
25
25
  - command: "/god-reconstruct"
@@ -20,6 +20,6 @@ sequences:
20
20
  description: "Greenfield idea to production with one command"
21
21
  steps:
22
22
  - command: "/god-mode"
23
- why: "Full autonomous arc, idea to hardened production"
23
+ why: "Full autonomous project run, idea to hardened production"
24
24
 
25
25
  default-sequence: default
@@ -24,6 +24,6 @@ sequences:
24
24
  - command: "/god-init"
25
25
  why: "Commit to clarified framing"
26
26
  - command: "/god-mode"
27
- why: "Run autonomous arc"
27
+ why: "Run autonomous project run"
28
28
 
29
29
  default-sequence: default
@@ -62,6 +62,13 @@ Warnings:
62
62
 
63
63
  Infos: 90 missing-recommended-section suggestions across agents
64
64
  (run with --json for full list)
65
+
66
+ Proposition:
67
+ 1. Implement partial: fix the highest-priority error or warning
68
+ 2. Implement complete: /god-agent-audit --fix for info-level placeholder sections only
69
+ 3. Discuss more: /god-discuss agent contract findings
70
+ 4. Inspect details: /god-agent-audit --json
71
+ Recommended: [one option and why]
65
72
  ```
66
73
 
67
74
  ## Backward-compatibility promise
@@ -45,4 +45,11 @@ Suggested next:
45
45
  /god-reconstruct - reverse-engineer planning artifacts from this code
46
46
  /god-tech-debt - assess and prioritize debt revealed
47
47
  /god-feature - now safe to add new work with archaeology in hand
48
+
49
+ Proposition:
50
+ 1. Implement partial: /god-tech-debt for only the highest-risk areas
51
+ 2. Implement complete: /god-reconstruct then /god-audit for full brownfield alignment
52
+ 3. Discuss more: /god-discuss the open tribal-knowledge questions
53
+ 4. Run God Mode: /god-mode only after reconstruction or audit makes the state clear
54
+ Recommended: /god-reconstruct when planning artifacts are missing or stale.
48
55
  ```
@@ -23,7 +23,7 @@ Spawn the **god-auditor** agent in a fresh context via Task tool.
23
23
  When invoked by `brownfield-arc` or `bluefield-arc` with
24
24
  `mode: greenfield-simulation`, spawn god-auditor with the current project
25
25
  evidence and ask it to compare the repo or org constraints against the
26
- canonical Godpowers greenfield arc. The agent writes
26
+ canonical Godpowers greenfield project run. The agent writes
27
27
  `.godpowers/audit/GREENFIELD-SIMULATION.md`.
28
28
 
29
29
  This mode builds nothing and rewrites no planning artifacts. It exists so
@@ -52,6 +52,16 @@ After god-auditor returns:
52
52
  1. Verify AUDIT-REPORT.md exists on disk
53
53
  2. Display the summary table to the user
54
54
  3. If any artifact scored below 80%: suggest re-running the failing tier
55
+ 4. End with a proposition block:
56
+
57
+ ```
58
+ Proposition:
59
+ 1. Implement partial: [rerun the single lowest-scoring tier or fix the top finding]
60
+ 2. Implement complete: [rerun all failing tiers in priority order]
61
+ 3. Discuss more: /god-discuss audit remediation plan
62
+ 4. Run God Mode: /god-mode only when the audit has no blocking findings
63
+ Recommended: [one option and why]
64
+ ```
55
65
 
56
66
  ## Output Format
57
67
 
@@ -89,6 +89,19 @@ Suggested next:
89
89
  2. Accept the 2 new todos: /god-check-todos
90
90
  ```
91
91
 
92
+ ## Proposition Closeout
93
+
94
+ End every context scan report with a proposition block:
95
+
96
+ ```
97
+ Proposition:
98
+ 1. Implement partial: [first suggested action, usually /god-locate or /god-repair]
99
+ 2. Implement complete: resolve all error and warn drifts, then rerun /god-context-scan
100
+ 3. Discuss more: /god-discuss context drift
101
+ 4. Inspect status: /god-status after re-orientation
102
+ Recommended: [one option and why]
103
+ ```
104
+
92
105
  ## What this does NOT do
93
106
 
94
107
  - Does not modify state. Read-only.
@@ -132,7 +132,7 @@ Done. AI tools will no longer see Godpowers context on this project.
132
132
 
133
133
  ## Auto-refresh
134
134
 
135
- When `/god-sync` runs (post-arc, or any sync), `god-updater` calls this skill
135
+ When `/god-sync` runs (after a project run, or any sync), `god-updater` calls this skill
136
136
  with `refresh` to keep `AGENTS.md` content aligned with the latest project
137
137
  state (mode, scale, completed tiers, active artifacts).
138
138
 
@@ -50,7 +50,7 @@ required, record the single access bundle in
50
50
  requests staging or final sign-off begins.
51
51
 
52
52
  The single access bundle must be incremental. Do not ask for
53
- `STAGING_APP_URL` mid-arc unless the user requested deployed staging. At final
53
+ `STAGING_APP_URL` mid-run unless the user requested deployed staging. At final
54
54
  sign-off or explicit staging, ask for the smallest next item needed to run the
55
55
  next command. If no live target URL is known, ask only for
56
56
  `STAGING_APP_URL=<staging-origin>` and the exact smoke command that will run.
@@ -65,6 +65,14 @@ Transitive (depth 2):
65
65
 
66
66
  Recommendation: Run /god-design with this change to apply.
67
67
  god-design-reviewer will gate via two-stage review.
68
+
69
+ Proposition:
70
+ 1. Implement partial: /god-design for only the lowest-risk token or component change
71
+ 2. Implement complete: /god-design with the full proposed change
72
+ 3. Discuss more: /god-discuss the design tradeoff before changing files
73
+ 4. Run God Mode: /god-mode only if this design change is part of a wider project run
74
+ Recommended: apply the smallest design change first when severity is warning
75
+ or higher.
68
76
  ```
69
77
 
70
78
  ## When this skill helps
@@ -67,4 +67,11 @@ Key findings:
67
67
  - [term or ambiguity resolved]
68
68
 
69
69
  Suggested next: [the planning command this discussion was for]
70
+
71
+ Proposition:
72
+ 1. Implement partial: [smallest safe slice or planning command]
73
+ 2. Implement complete: [full command that consumes this brief]
74
+ 3. Discuss more: /god-discuss [specific unresolved question]
75
+ 4. Run God Mode: /god-mode [scope] if the user wants autonomous execution
76
+ Recommended: [one option and why]
70
77
  ```
@@ -65,6 +65,23 @@ Suggested next steps:
65
65
  2. /god-restore (review trash)
66
66
  ```
67
67
 
68
+ ## Proposition Closeout
69
+
70
+ End every human-readable doctor report with a proposition block:
71
+
72
+ ```
73
+ Proposition:
74
+ 1. Implement partial: [safest single fix or diagnostic follow-up]
75
+ 2. Implement complete: /god-doctor --fix when all proposed fixes are safe categories
76
+ 3. Discuss more: /god-discuss [highest-risk warning or unclear repair]
77
+ 4. Inspect status: /god-status after repair
78
+ Recommended: [one option and why it is safe]
79
+ ```
80
+
81
+ If the report contains errors that need manual repair, do not recommend
82
+ `/god-doctor --fix` as complete. Recommend the highest-priority manual repair
83
+ or `/god-repair` instead.
84
+
68
85
  ## Subcommands
69
86
 
70
87
  ### `/god-doctor`
@@ -47,4 +47,12 @@ Idea explored. Suggested framing:
47
47
 
48
48
  Suggested next: /god-init (commit to this framing) or /god-explore again
49
49
  (if you want to try another angle)
50
+
51
+ Proposition:
52
+ 1. Implement partial: /god-init with this framing and stop after PRD
53
+ 2. Implement complete: /god-mode with this framing for the full project run
54
+ 3. Discuss more: /god-explore again with the weakest assumption
55
+ 4. Run God Mode: /god-mode if the user is ready to build now
56
+ Recommended: /god-init when the framing feels stable enough to turn into
57
+ requirements.
50
58
  ```
@@ -15,7 +15,7 @@ Capture institutional knowledge from a completed phase or milestone.
15
15
  ## When to use
16
16
 
17
17
  - After completing a milestone (Now -> Done)
18
- - After a successful /god-mode arc
18
+ - After a successful /god-mode project run
19
19
  - Before a /god-postmortem (different focus: that's for incidents)
20
20
 
21
21
  ## Process
@@ -1,7 +1,7 @@
1
1
  ---
2
2
  name: god-feature
3
3
  description: |
4
- Add a feature to an existing project. Skips the full-arc setup (no /god-init,
4
+ Add a feature to an existing project. Skips the full project-run setup (no /god-init,
5
5
  no repo scaffolding, no /god-stack). Mini-PRD for the feature, optional
6
6
  micro-architecture, build, deploy via existing pipeline, harden new code.
7
7