godpowers 0.15.0

package/templates/POSTMORTEM.md

@@ -0,0 +1,88 @@
+# Incident <ID>: [Short title]
+
+> Blameless postmortem. Frame everything as systems, not people.
+> Action items must have owner, due date, and observable success criterion.
+
+Date: [ISO 8601]
+Severity: SEV-[1/2/3]
+Duration: [from detection to recovery]
+Reporter: [who filed]
+
+## Summary
+
+[2-3 sentence executive summary. What happened, what was the impact, what
+was the fix.]
+
+## Impact
+
+- Users affected: [number, percentage]
+- Revenue impact: [if applicable, with currency]
+- Data loss: [yes/no, scope]
+- SLO breach: [which SLO, how much budget consumed]
+
+## Timeline
+
+| Time (UTC) | Event |
+|------------|-------|
+| HH:MM | First symptom appeared (according to [source]) |
+| HH:MM | Detected by [alert / user report / monitoring] |
+| HH:MM | Acknowledged by [who] |
+| HH:MM | [Investigation step] |
+| HH:MM | Cause identified |
+| HH:MM | Fix deployed |
+| HH:MM | System fully recovered |
+
+Acknowledge gaps: any unknown intervals must be marked "unknown" with a note
+on what evidence would resolve them.
+
+## Root Cause
+
+[Specific technical root cause. Walk through the 5 whys or causal chain.
+Don't stop at the first plausible cause.]
+
+## Class of Bug
+
+[The broader pattern. The class names what other instances could occur if
+left unaddressed.]
+
+Example: Instance was "auth endpoint returned 500 for emails with `+`".
+Class is "input validation regexes don't cover the email RFC's full charset".
+
+## What Went Well
+
+- [Specific thing that worked, with evidence]
+- [Another specific thing]
+
+## What Didn't
+
+- [Specific gap, with evidence] [DECISION]
+- [Another gap] [HYPOTHESIS] (we believe but haven't confirmed)
+
+## Action Items
+
+| Action | Owner | Due | Priority | Success Criterion |
+|--------|-------|-----|----------|-------------------|
+| [Specific change tied to class-of-bug] | [name] | [date] | P0 | [observable] |
+| [Specific change] | [name] | [date] | P1 | [observable] |
+
+Priority: P0 = next sprint, P1 = next quarter, P2 = backlog.
+
+## Runbook Updates
+
+- Updated: [path] - [what changed and why]
+- Created: [path] - [why it didn't exist]
+- No update: [path] - [why it was already adequate]
+
+---
+
+## Have-Nots Checklist
+
+- [ ] Action items are specific (not "communicate better")
+- [ ] Every action item has named owner
+- [ ] Every action item has specific due date
+- [ ] Every action item has observable success criterion
+- [ ] Root cause goes beyond the symptom
+- [ ] Class-of-bug identified
+- [ ] Phrasing is blameless (systems, not people)
+- [ ] Runbooks updated where gaps existed
+- [ ] Timeline gaps acknowledged explicitly

package/templates/PRD.md

@@ -0,0 +1,80 @@
+# Product Requirements Document
+
+> Every sentence below MUST be labeled `[DECISION]`, `[HYPOTHESIS]`, or `[OPEN QUESTION]`.
+> Every claim MUST fail the substitution test (swap competitor name in, sentence breaks).
+
+## Problem Statement
+
+[DECISION] [Specific problem that real people have today.]
+
+[HYPOTHESIS] [What you believe is true but haven't proven.]
+
+[OPEN QUESTION] [Unresolved item] -- Owner: [name] -- Due: [date]
+
+## Target Users
+
+[DECISION] Primary: [Specific persona, not "developers" or "users". Include context: their role, what they care about, what they do today.]
+
+[DECISION] Secondary: [If applicable. Specific.]
+
+## Success Metrics
+
+[DECISION] [Number] [unit] within [timeline], measured by [method].
+- Example: 50 daily active users within 30 days of launch, measured via analytics
+
+[DECISION] [Second metric with number, timeline, measurement.]
+
+## Functional Requirements
+
+### MUST (V1 launch blockers)
+- [DECISION] [Requirement] -- Acceptance: [observable criterion]
+- [DECISION] [Requirement] -- Acceptance: [observable criterion]
+
+### SHOULD (V1 if time permits)
+- [HYPOTHESIS] [Requirement] -- Acceptance: [criterion] -- Validation: [how we'll test if this matters]
+
+### COULD (post-V1)
+- [HYPOTHESIS] [Requirement]
+
+## Non-Functional Requirements
+
+| Category | Requirement | Source |
+|----------|-------------|--------|
+| Latency | p99 < [N]ms for [operation] | [DECISION] |
+| Availability | [N]% uptime over 30 days | [DECISION] |
+| Scale | Support [N] concurrent users / [N] requests/sec | [HYPOTHESIS] |
+| Security | [Auth model, data classification, compliance] | [DECISION] |
+
+## Scope and No-Gos
+
+### In scope for V1
+- [List]
+
+### Explicitly NOT in scope
+- [DECISION] [What we are not building. This section CANNOT be empty.]
+- [DECISION] [Another no-go.]
+
+## Appetite
+
+[DECISION] Time budget: [N weeks/months from start to launch]
+[DECISION] Resource budget: [team size, dollar budget if applicable]
+[DECISION] Technical constraints: [must use X, cannot use Y]
+
+## Open Questions
+
+| Question | Owner | Due Date | Resolution |
+|----------|-------|----------|------------|
+| [Question] | [Name] | [Date] | [Empty until resolved] |
+
+---
+
+## Have-Nots Checklist
+
+Before declaring done, verify:
+- [ ] No sentence is unlabeled
+- [ ] Problem statement fails substitution test
+- [ ] Target user is specific (not "developers")
+- [ ] Every success metric has a number AND timeline
+- [ ] Every requirement has acceptance criteria
+- [ ] No-gos section is non-empty
+- [ ] Every open question has owner AND due date

package/templates/PROGRESS.md

@@ -0,0 +1,49 @@
+# Godpowers Progress
+
+> Disk-authoritative. This file reflects what's actually on disk, not what
+> the agent believes happened.
+
+Mode: [A: greenfield | B: gap-fill | C: audit | D: multi-repo]
+Scale: [trivial | small | medium | large | enterprise]
+Started: [ISO 8601 timestamp]
+Last updated: [ISO 8601 timestamp]
+
+## Project Description
+
+[One paragraph from /god-init]
+
+## Tier Status
+
+| Tier | Sub-step | Status | Artifact | Updated |
+|------|----------|--------|----------|---------|
+| 0 | Orchestration | done | .godpowers/PROGRESS.md | [timestamp] |
+| 1 | PRD | pending | -- | -- |
+| 1 | Architecture | pending | -- | -- |
+| 1 | Roadmap | pending | -- | -- |
+| 1 | Stack | pending | -- | -- |
+| 2 | Repo | pending | -- | -- |
+| 2 | Build | pending | -- | -- |
+| 3 | Deploy | pending | -- | -- |
+| 3 | Observe | pending | -- | -- |
+| 3 | Launch | pending | -- | -- |
+| 3 | Harden | pending | -- | -- |
+
+Valid statuses: `pending`, `in-flight`, `done`, `skipped`, `imported`, `failed`, `re-invoked`
+
+## Decisions Log
+
+[Append-only log of significant decisions made during the run]
+
+- [timestamp] PRD: Target user clarified as "solo SaaS founders, $1k-$10k MRR"
+- [timestamp] ARCH: Chose monolith over microservices (ADR-001)
+
+## Pause Log
+
+[Append-only record of every pause and resolution]
+
+- [timestamp] Stack pause: TypeScript vs Python -> User chose TypeScript
+- [timestamp] Critical finding pause: Auth bypass -> User chose Fix Now (Option A)
+
+## YOLO Decisions (when --yolo is active)
+
+[If --yolo flag was used, all auto-picked defaults are logged here for review]

package/templates/ROADMAP.md

@@ -0,0 +1,47 @@
+# Roadmap
+
+> Every milestone goal is substitution-tested. Every completion gate is
+> observable. No day-level precision without capacity input.
+
+## Now (currently building, committed)
+
+### Milestone 1: [Substitution-tested goal]
+- **Goal**: [What users can DO when this ships. Specific.]
+- **Completion gate**: [Observable criterion. Not "feels done".]
+- **Size**: S / M / L
+- **Depends on**: None (or list)
+- **Features (from PRD)**:
+  - [PRD requirement reference]
+  - [PRD requirement reference]
+
+### Milestone 2: [Substitution-tested goal]
+- **Goal**: ...
+- **Completion gate**: ...
+- **Size**: S / M / L
+- **Depends on**: Milestone 1
+- **Features**: ...
+
+## Next (planned, flexible)
+
+### Milestone 3: [Goal]
+- **Goal**: ...
+- **Completion gate**: ...
+- **Size**: ...
+- **Depends on**: Milestone 2
+
+## Later (intent, not commitment)
+
+### Theme: [Substitution-tested theme name]
+- [Possible directions; may change based on what we learn]
+
+---
+
+## Have-Nots Checklist
+
+- [ ] Every milestone goal fails substitution test
+- [ ] Every completion gate is observable (not subjective)
+- [ ] No feature appears that isn't in the PRD
+- [ ] Milestones are NOT all the same size (real prioritization)
+- [ ] Dependency edges are explicit
+- [ ] No day-level precision without capacity input
+- [ ] Later section has at least one theme

package/templates/SPIKE.md

@@ -0,0 +1,72 @@
+# Spike: [The Question]
+
+> Time-boxed research. Findings only, with evidence. Spike code is throwaway.
+> Do NOT merge to main.
+
+Date: [ISO 8601]
+Time-box: [N hours, default 1 day]
+Time spent: [actual hours]
+Status: [resolved / inconclusive / time-boxed-out]
+
+## Question
+
+[Exact question, one sentence. Specific enough to be answerable.]
+
+Bad: "Should we use GraphQL?"
+Good: "Can our existing REST endpoints be auto-generated as GraphQL with
+sub-100ms p95 latency for our top 5 query patterns?"
+
+## Approach
+
+[What you built and why. Keep it minimal.]
+
+## Findings
+
+### What works
+- [Specific finding] [DECISION]
+  - Evidence: [numbers, code, test output]
+
+### What doesn't
+- [Specific finding] [DECISION]
+  - Evidence: [numbers, code, test output]
+
+### Open
+- [Things you didn't get to] [OPEN QUESTION]
+  - Owner: [user]
+  - Due: [date if you plan to revisit]
+
+## Evidence
+
+[Numbers. Screenshots. Code excerpts. Test results.]
+
+NOT prose claims without backing.
+
+```
+[example: latency measurements, error rates, code snippets]
+```
+
+## Recommendation
+
+[ONE recommendation, [DECISION] tagged.]
+
+OR: "Inconclusive. Suggested follow-up spike: <narrower question>."
+
+## Code Location
+
+[Where the spike code lives. Note it's throwaway.]
+
+## Reversibility
+
+This spike code MUST NOT be merged to main. If the recommendation is to
+proceed, route to /god-feature with this recommendation. The feature
+implementation rewrites cleanly using production patterns.
+
+---
+
+## Have-Nots Checklist
+
+- [ ] Built a proof, not a feature
+- [ ] All claims have evidence (numbers, code, test output)
+- [ ] Made a recommendation (or honest "inconclusive" with follow-up)
+- [ ] Time-box respected (or escalated)
+- [ ] Spike code clearly marked throwaway

package/templates/STACK-DECISION.md

@@ -0,0 +1,61 @@
+# Stack Decision
+
+> Every choice has a flip point. Every choice has a lock-in cost.
+
+## Summary
+
+| Category | Choice | Score | Lock-in | Flip Point |
+|----------|--------|-------|---------|------------|
+| Language/Runtime | [Choice] | [N.N/10] | Low/Med/High | [Condition that reverses this] |
+| Web Framework | [Choice] | [N.N/10] | Low/Med/High | [Condition] |
+| Database | [Choice] | [N.N/10] | Low/Med/High | [Condition] |
+| Cache | [Choice] | [N.N/10] | Low/Med/High | [Condition] |
+| Hosting | [Choice] | [N.N/10] | Low/Med/High | [Condition] |
+| Auth | [Choice] | [N.N/10] | Low/Med/High | [Condition] |
+| Observability | [Choice] | [N.N/10] | Low/Med/High | [Condition] |
+| CI/CD | [Choice] | [N.N/10] | Low/Med/High | [Condition] |
+
+## Detailed Decisions
+
+### Language/Runtime: [Choice]
+- **Candidates evaluated**: [Choice], [Alternative 1], [Alternative 2]
+- **Scores**: [Choice] [N.N] / [Alt1] [N.N] / [Alt2] [N.N]
+- **Why this one**: [Specific rationale tied to ARCH choices, NFRs from PRD]
+- **Flip point**: [Specific condition under which we reverse this]
+  - Example: "If team grows past 5 engineers and the majority prefer Python"
+- **Lock-in cost**: Low / Medium / High
+  - **What switching requires**: [Concrete description]
+  - **Estimated effort**: [Person-weeks/months]
+
+### Web Framework: [Choice]
+[Same structure]
+
+### Database: [Choice]
+[Same structure]
+
+[Continue for each category]
+
+## Pairing Compatibility
+
+Verified compatible pairings:
+- [Choice A] works with [Choice B]: [evidence]
+- [Choice C] works with [Choice D]: [evidence]
+
+Flagged concerns:
+- [Any incompatibilities or version constraints]
+
+## High Lock-In Watch List
+
+Choices marked "High" lock-in cost with flip points likely within 6 months:
+- [None] OR [List with mitigation plan]
+
+---
+
+## Have-Nots Checklist
+
+- [ ] Every category has a choice with rationale
+- [ ] Every choice has a flip point
+- [ ] Every choice has a lock-in cost classification
+- [ ] No "best practice" rationale without specific tie to ARCH
+- [ ] Pairing compatibility verified
+- [ ] High lock-in choices acknowledged or mitigated

package/workflows/audit-only.yaml

@@ -0,0 +1,22 @@
+# STATUS: Authoritative (v0.14+). Read by lib/workflow-runner.js for plan
+# computation. The orchestrator agent (agents/god-orchestrator.md) dispatches
+# the agents in the order this file specifies. To preview without running,
+# use `/god-mode --workflow=<name> --plan`.
+#
+apiVersion: godpowers/v1
+kind: Workflow
+metadata:
+  name: audit-only
+  version: 1.0.0
+  description: |
+    Score existing artifacts against have-nots. Build nothing.
+    Run by /god-audit.
+
+on: [/god-audit]
+
+jobs:
+  audit-score:
+    tier: 0
+    uses: god-auditor@^1.0.0
+    with:
+      mode: full-audit

package/workflows/bluefield-arc.yaml

@@ -0,0 +1,87 @@
+# STATUS: Authoritative (v0.14+). Read by lib/workflow-runner.js for plan
+# computation. The orchestrator agent (agents/god-orchestrator.md) dispatches
+# the agents in the order this file specifies. To preview without running,
+# use `/god-mode --workflow=<name> --plan`.
+#
+apiVersion: godpowers/v1
+kind: Workflow
+metadata:
+  name: bluefield-arc
+  version: 1.0.0
+  description: |
+    Bluefield: new code, existing org context. Loads org standards first,
+    then runs full-arc with all decisions constrained by org context.
+
+on: [/god-bluefield, /god-mode --bluefield]
+
+jobs:
+  org-context:
+    tier: 0
+    uses: god-org-context-loader@^1.0.0
+
+  prd:
+    tier: 1
+    needs: org-context
+    uses: god-pm@^1.0.0
+
+  arch:
+    tier: 1
+    needs: prd
+    uses: god-architect@^1.0.0
+    with:
+      respect-org-context: true
+
+  roadmap:
+    tier: 1
+    needs: arch
+    uses: god-roadmapper@^1.0.0
+
+  stack:
+    tier: 1
+    needs: arch
+    uses: god-stack-selector@^1.0.0
+    with:
+      respect-org-context: true
+
+  repo:
+    tier: 2
+    needs: stack
+    uses: god-repo-scaffolder@^1.0.0
+
+  build:
+    tier: 2
+    needs: [roadmap, repo]
+    uses: god-planner@^1.0.0
+    review:
+      stages:
+        - uses: god-spec-reviewer@^1.0.0
+        - uses: god-quality-reviewer@^1.0.0
+    on-pass: commit-atomic
+
+  deploy:
+    tier: 3
+    needs: build
+    uses: god-deploy-engineer@^1.0.0
+    with:
+      use-org-platform: true
+
+  observe:
+    tier: 3
+    needs: deploy
+    uses: god-observability-engineer@^1.0.0
+    with:
+      use-org-stack: true
+
+  harden:
+    tier: 3
+    needs: build
+    uses: god-harden-auditor@^1.0.0
+    with:
+      check-org-security-standards: true
+    blocks-on:
+      - critical-finding: pause
+
+  launch:
+    tier: 3
+    needs: harden
+    uses: god-launch-strategist@^1.0.0

package/workflows/brownfield-arc.yaml

@@ -0,0 +1,44 @@
+# STATUS: Authoritative (v0.14+). Read by lib/workflow-runner.js for plan
+# computation. The orchestrator agent (agents/god-orchestrator.md) dispatches
+# the agents in the order this file specifies. To preview without running,
+# use `/god-mode --workflow=<name> --plan`.
+#
+apiVersion: godpowers/v1
+kind: Workflow
+metadata:
+  name: brownfield-arc
+  version: 1.0.0
+  description: |
+    Brownfield: existing codebase, understand before changing. Archaeology +
+    reconstruction + debt assessment, then proceed with normal arc workflows.
+
+on: [/god-brownfield, /god-mode --brownfield]
+
+jobs:
+  archaeology:
+    tier: 0
+    uses: god-archaeologist@^1.0.0
+
+  reconstruct:
+    tier: 1
+    needs: archaeology
+    uses: god-reconstructor@^1.0.0
+
+  debt-assess:
+    tier: 0
+    needs: archaeology
+    uses: god-debt-assessor@^1.0.0
+
+  audit:
+    tier: 0
+    needs: [reconstruct, debt-assess]
+    uses: god-auditor@^1.0.0
+    with:
+      mode: full-audit-on-reconstructed
+
+  steady-state-handoff:
+    tier: 0
+    needs: audit
+    uses: god-orchestrator@^1.0.0
+    with:
+      action: brownfield-to-steady

package/workflows/deps-audit.yaml

@@ -0,0 +1,56 @@
+# STATUS: Authoritative (v0.14+). Read by lib/workflow-runner.js for plan
+# computation. The orchestrator agent (agents/god-orchestrator.md) dispatches
+# the agents in the order this file specifies. To preview without running,
+# use `/god-mode --workflow=<name> --plan`.
+#
+apiVersion: godpowers/v1
+kind: Workflow
+metadata:
+  name: deps-audit
+  version: 1.0.0
+  description: |
+    Audit and update dependencies. Critical CVEs first, batched patches,
+    per-package minor updates, majors deferred to /god-upgrade.
+    Run by /god-update-deps.
+
+on: [/god-update-deps]
+
+jobs:
+  deps-audit-scan:
+    tier: 0
+    uses: god-deps-auditor@^1.0.0
+    with:
+      template: DEPS-AUDIT.md
+
+  deps-patch-batch:
+    tier: 2
+    needs: deps-audit-scan
+    uses: god-executor@^1.0.0
+    with:
+      mode: batch-patch-updates
+    review:
+      stages:
+        - uses: god-quality-reviewer@^1.0.0
+    on-pass: commit-atomic
+
+  deps-minor-per-package:
+    tier: 2
+    needs: deps-audit-scan
+    uses: god-executor@^1.0.0
+    with:
+      mode: per-package-minor
+    per: slice
+    review:
+      stages:
+        - uses: god-quality-reviewer@^1.0.0
+          on-fail: rollback
+    on-pass: commit-atomic
+
+  deps-major-handoff:
+    tier: 0
+    needs: deps-audit-scan
+    uses: god-orchestrator@^1.0.0
+    with:
+      action: route-to
+      target: /god-upgrade
+      per: major-version-bump

package/workflows/docs-arc.yaml

@@ -0,0 +1,22 @@
+# STATUS: Authoritative (v0.14+). Read by lib/workflow-runner.js for plan
+# computation. The orchestrator agent (agents/god-orchestrator.md) dispatches
+# the agents in the order this file specifies. To preview without running,
+# use `/god-mode --workflow=<name> --plan`.
+#
+apiVersion: godpowers/v1
+kind: Workflow
+metadata:
+  name: docs-arc
+  version: 1.0.0
+  description: |
+    Verify and update documentation against code.
+    Run by /god-docs.
+
+on: [/god-docs]
+
+jobs:
+  docs-update:
+    tier: 1
+    uses: god-docs-writer@^1.0.0
+    with:
+      template: DOCS-UPDATE-LOG.md

package/workflows/feature-arc.yaml

@@ -0,0 +1,59 @@
+# STATUS: Authoritative (v0.14+). Read by lib/workflow-runner.js for plan
+# computation. The orchestrator agent (agents/god-orchestrator.md) dispatches
+# the agents in the order this file specifies. To preview without running,
+# use `/god-mode --workflow=<name> --plan`.
+#
+apiVersion: godpowers/v1
+kind: Workflow
+metadata:
+  name: feature-arc
+  version: 1.0.0
+  description: |
+    Add a feature to an existing project. Skips full-arc setup.
+    Run by /god-feature.
+
+on: [/god-feature]
+
+jobs:
+  feature-prd:
+    tier: 1
+    uses: god-pm@^1.0.0
+    with:
+      mode: feature-only
+      template: PRD.md
+
+  feature-arch:
+    tier: 1
+    needs: feature-prd
+    uses: god-architect@^1.0.0
+    with:
+      mode: delta-only
+    skip-when: existing-arch-covers-feature
+
+  feature-build:
+    tier: 2
+    needs: feature-prd
+    uses: god-planner@^1.0.0
+    review:
+      stages:
+        - uses: god-spec-reviewer@^1.0.0
+          on-fail: rollback
+        - uses: god-quality-reviewer@^1.0.0
+          on-fail: rollback
+    on-pass: commit-atomic
+
+  feature-harden:
+    tier: 3
+    needs: feature-build
+    uses: god-harden-auditor@^1.0.0
+    with:
+      mode: scope-to-new-code
+    blocks-on:
+      - critical-finding: pause
+
+  feature-soft-launch:
+    tier: 3
+    needs: feature-harden
+    uses: god-launch-strategist@^1.0.0
+    with:
+      mode: feature-flag-rollout