@devran-ai/kit 4.1.0

package/.agent/workflows/create.md

@@ -0,0 +1,85 @@
+---
+description: Create new features, components, or modules from scratch.
+version: 2.1.0
+sdlc-phase: build
+skills: [app-builder, clean-code]
+commit-types: [feat]
+---
+
+# /create — Scaffold New Features
+
+> **Trigger**: `/create [description]`
+> **Lifecycle**: Build — after `/plan` approval
+
+> Standards: See `rules/workflow-standards.md`
+
+---
+
+## Critical Rules
+
+1. Follow existing patterns — scan codebase before writing
+2. No orphan code — every file must be imported/referenced/routed
+3. Tests required for all new features
+4. Stack-agnostic detection from config files
+5. User approval for scaffolds creating >5 files
+6. Document all exported public APIs
+
+---
+
+## Steps
+
+// turbo
+1. **Clarify Requirements** — component type, acceptance criteria, constraints
+
+// turbo
+2. **Detect Stack** — auto-detect from config files, identify framework/conventions
+
+// turbo
+3. **Analyze Patterns** — find similar modules, naming conventions, import patterns, reusable utils
+
+4. **Present Plan** (for >5 files) — show file structure, integration points. Wait for approval.
+
+5. **Implement** — follow detected conventions, SOLID principles, wire up imports/routes
+
+6. **Add Tests** — unit, integration, E2E as applicable
+
+7. **Document** — JSDoc/docstrings, README updates, usage examples
+
+---
+
+## Output Template
+
+```markdown
+## Create: [Feature]
+
+- **Stack**: [language] + [framework]
+- **Files Created**: [list with purposes]
+- **Integration**: [how it connects]
+- **Tests**: [what's covered]
+
+**Next**: `/test` or `/preview`
+```
+
+---
+
+## Governance
+
+**PROHIBITED:** Creating without checking patterns · wrong-stack scaffolding · orphan files · skipping tests
+
+**REQUIRED:** Stack detection · pattern analysis · user approval for >5 files · test coverage · integration
+
+---
+
+## Completion Criteria
+
+- [ ] Stack detected, patterns analyzed
+- [ ] Files created and integrated
+- [ ] Tests written and passing
+- [ ] Documentation added
+
+---
+
+## Related Resources
+
+- **Previous**: `/plan` · **Next**: `/test` · `/preview`
+- **Skill**: `.agent/skills/app-builder/SKILL.md`

package/.agent/workflows/debug.md

@@ -0,0 +1,83 @@
+---
+description: Systematic debugging workflow. Activates DEBUG mode for problem investigation.
+version: 2.1.0
+sdlc-phase: reactive
+skills: [debugging-strategies]
+commit-types: [fix]
+---
+
+# /debug — Systematic Problem Investigation
+
+> **Trigger**: `/debug [issue description]`
+> **Lifecycle**: Reactive — any SDLC phase
+
+> Standards: See `rules/workflow-standards.md`
+
+---
+
+## Critical Rules
+
+1. Root cause required — never fix without understanding why
+2. No guessing — form hypotheses, test systematically
+3. Prevention mandatory — every fix includes recurrence prevention
+4. Preserve evidence before changing anything
+5. Minimal changes — fix only what's broken
+
+---
+
+## Steps
+
+// turbo
+1. **Gather Info** — exact error/stack trace, reproduction steps, expected vs actual
+
+// turbo
+2. **Environment** — OS, runtime versions, recent git changes, deps, config
+
+// turbo
+3. **Hypotheses** — list 3+ causes ordered by likelihood
+
+// turbo
+4. **Investigate** — test each hypothesis, check logs/data/state, eliminate definitively
+
+5. **Fix** — minimal fix for root cause, verify resolution, confirm no regressions
+
+6. **Prevent** — add tests, validation, guardrails, document root cause
+
+---
+
+## Output Template
+
+```markdown
+## Debug: [Issue]
+
+1. **Symptom**: [error/behavior]
+2. **Root Cause**: [explanation]
+3. **Fix**: [changes applied]
+4. **Prevention**: [tests/guardrails added]
+
+**Next**: `/test` for regression check.
+```
+
+---
+
+## Governance
+
+**PROHIBITED:** Fixing without root cause · random guessing · modifying production without rollback
+
+**REQUIRED:** Hypothesis testing · root cause documentation · prevention measures · regression verification
+
+---
+
+## Completion Criteria
+
+- [ ] Issue reproduced and documented
+- [ ] Root cause identified with evidence
+- [ ] Fix applied and verified
+- [ ] Prevention measures implemented
+
+---
+
+## Related Resources
+
+- **Next**: `/test`
+- **Skill**: `.agent/skills/debugging-strategies/SKILL.md`

package/.agent/workflows/deploy.md

@@ -0,0 +1,114 @@
+---
+description: Production deployment with pre-flight checks, execution, and verification.
+version: 2.1.0
+sdlc-phase: ship
+skills: [deployment-procedures]
+commit-types: [chore, fix]
+---
+
+# /deploy — Production Deployment
+
+> **Trigger**: `/deploy [sub-command]`
+> **Lifecycle**: Ship — after `/pr` is merged
+
+> Standards: See `rules/workflow-standards.md`
+
+> [!CAUTION]
+> Deployment impacts production users and consumes platform credits. Never deploy untested code.
+
+---
+
+## Critical Rules
+
+1. Rollback plan required before deploying
+2. No test deploys to production — use preview/staging
+3. Pre-flight must pass before deployment
+4. Health check mandatory after deployment
+5. No secrets in deployment logs
+
+---
+
+## Scope Filter
+
+Deploy only production-impacting changes: `apps/`, `docker/`, infra config. Never deploy docs-only or `.agent/` changes.
+
+---
+
+## Argument Parsing
+
+| Command | Action |
+| :--- | :--- |
+| `/deploy` | Interactive wizard |
+| `/deploy check` | Pre-flight only |
+| `/deploy preview` | Deploy to staging |
+| `/deploy production` | Deploy to production |
+| `/deploy rollback` | Rollback to previous |
+
+---
+
+## Steps
+
+// turbo
+1. **Pre-Flight** — tsc, lint, tests, security audit, build, environment check
+
+// turbo
+2. **Scope Verification** — check changed files against scope filter
+
+3. **Rollback Plan** — document current version, verify rollback command, check migration reversibility
+
+4. **Deploy** — build, deploy to target, monitor progress
+
+// turbo
+5. **Health Check** — API responding, database connected, services healthy, no error spikes
+
+6. **Post-Deploy** — document version/SHA, update tracking, notify stakeholders
+
+---
+
+## Platform Support
+
+| Platform | Command | Auto-detect |
+| :--- | :--- | :--- |
+| Vercel | `vercel --prod` | Next.js |
+| Railway | `railway up` | NestJS, API |
+| Expo EAS | `eas build` | React Native |
+
+---
+
+## Output Template
+
+```markdown
+## Deployment Complete
+
+- **Version**: [SHA/tag]
+- **Environment**: [target]
+- **Health**: All checks passing
+- **Rollback**: `/deploy rollback` to [previous]
+
+**Next**: `/status` for monitoring.
+```
+
+---
+
+## Governance
+
+**PROHIBITED:** Deploying without `/review` · production for testing · docs-only deploys · skipping rollback plan
+
+**REQUIRED:** Pre-flight passing · scope verification · rollback plan · health check · cost-conscious batching
+
+---
+
+## Completion Criteria
+
+- [ ] Pre-flight passed, scope verified
+- [ ] Rollback plan documented
+- [ ] Deployed, health check passed
+- [ ] Version documented
+
+---
+
+## Related Resources
+
+- **Previous**: `/pr` (merged) · `/preflight`
+- **Next**: `/status`
+- **Skill**: `.agent/skills/deployment-procedures/SKILL.md`

package/.agent/workflows/enhance.md

@@ -0,0 +1,85 @@
+---
+description: Add or update features in existing application. Iterative development.
+version: 2.1.0
+sdlc-phase: build
+skills: [clean-code, testing-patterns]
+commit-types: [feat, refactor]
+---
+
+# /enhance — Iterative Feature Development
+
+> **Trigger**: `/enhance [description]`
+> **Lifecycle**: Build — after `/plan` or ad-hoc for minor updates
+
+> Standards: See `rules/workflow-standards.md`
+
+---
+
+## Critical Rules
+
+1. Preserve existing functionality — never break what works
+2. Regression check mandatory after every change
+3. User approval for changes affecting >5 files
+4. Incremental approach — small, verifiable changes
+5. Follow existing conventions
+6. Document user-facing changes
+
+---
+
+## Steps
+
+// turbo
+1. **Understand State** — explore structure, review features/stack/conventions, identify relevant files
+
+// turbo
+2. **Impact Analysis** — affected files, dependencies, regression risk, breaking changes
+
+3. **Present Plan** (for >5 files) — show scope, affected areas, risk level. Wait for approval.
+
+4. **Implement** — follow existing patterns, apply incrementally
+
+// turbo
+5. **Regression Check** — tests, build, lint, type-check
+
+6. **Document** — update inline docs, README if user-facing, changelog if applicable
+
+---
+
+## Output Template
+
+```markdown
+## Enhancement: [Feature]
+
+| Action | File | Description |
+| :--- | :--- | :--- |
+| Modified/Created | `path` | [what] |
+
+- **Risk**: Low/Medium/High
+- **Regression**: tests/build/lint passing
+
+**Next**: `/test` or `/preview`
+```
+
+---
+
+## Governance
+
+**PROHIBITED:** Breaking existing functionality · large changes without impact analysis · skipping regression checks
+
+**REQUIRED:** Impact analysis · user approval for >5 files · regression check · documentation updates
+
+---
+
+## Completion Criteria
+
+- [ ] Impact analysis complete
+- [ ] Changes implemented following patterns
+- [ ] Regression check passed
+- [ ] Documentation updated
+
+---
+
+## Related Resources
+
+- **Previous**: `/plan` · **Next**: `/test` · `/preview`
+- **Skill**: `.agent/skills/clean-code/SKILL.md`

package/.agent/workflows/orchestrate.md

@@ -0,0 +1,106 @@
+---
+description: Multi-agent orchestration for complex tasks requiring multiple specialists.
+version: 2.1.0
+sdlc-phase: reactive
+agents: [planner, explorer-agent]
+skills: [parallel-agents, intelligent-routing]
+commit-types: [feat, refactor, fix]
+---
+
+# /orchestrate — Multi-Agent Coordination
+
+> **Trigger**: `/orchestrate [task description]`
+> **Lifecycle**: Reactive — complex multi-domain tasks at any SDLC phase
+
+> Standards: See `rules/workflow-standards.md`
+
+> [!CAUTION]
+> Coordinates multiple agents on same codebase. Ensure clear domain boundaries. Phase 2 requires explicit user approval.
+
+---
+
+## Critical Rules
+
+1. **2-Phase protocol** — always plan before implementing
+2. User approval gate between phases
+3. Every subagent receives full context (request, decisions, prior work)
+4. No agent conflicts — separate files or delineated domains
+5. Verification required after all agents complete
+
+---
+
+## Steps
+
+### PHASE 1: Planning (Sequential)
+
+// turbo
+1. **Analyze Domains** — identify all domains, map needed agents
+
+// turbo
+2. **Create Plan** — structured breakdown with execution order and dependencies
+
+3. **User Approval** — present plan, wait for explicit approval
+
+### PHASE 2: Implementation (After Approval)
+
+4. **Execute in Groups** — Foundation (data, security) → Core (app logic) → Quality (tests) → Operations (infra)
+
+5. **Context Passing** — every subagent gets: original request, decisions, prior agent work, plan state
+
+// turbo
+6. **Verification** — tests, lint, type-check, build
+
+---
+
+## Agent Selection
+
+| Domain | Agent(s) |
+| :--- | :--- |
+| Architecture | `architect`, `planner` |
+| Backend/DB | `backend-specialist`, `database-architect` |
+| Frontend | `frontend-specialist` |
+| Mobile | `mobile-developer` |
+| Security | `security-reviewer` |
+| Testing | `tdd-guide`, `e2e-runner` |
+| DevOps | `devops-engineer` |
+| Performance | `performance-optimizer` |
+| Code Quality | `refactor-cleaner`, `code-reviewer` |
+
+---
+
+## Output Template
+
+```markdown
+## Orchestration Complete
+
+### Agents Invoked
+| Agent | Domain | Summary |
+
+### Deliverables
+| Action | File | Agent |
+
+### Verification
+Tests / Build / Lint: status
+```
+
+---
+
+## Governance
+
+**PROHIBITED:** Skipping Phase 1 · Phase 2 without approval · agents without context · overlapping files · skipping verification
+
+**REQUIRED:** 2-Phase protocol · full context passing · domain boundaries · verification after completion
+
+---
+
+## Completion Criteria
+
+- [ ] Domains analyzed, plan approved
+- [ ] Agents executed with context
+- [ ] Verification passed
+
+---
+
+## Related Resources
+
+- **Skills**: `.agent/skills/parallel-agents/SKILL.md` · `.agent/skills/intelligent-routing/SKILL.md`

package/.agent/workflows/plan.md

@@ -0,0 +1,105 @@
+---
+description: Create implementation plan. Invokes planner agent for structured task breakdown.
+version: 2.2.0
+sdlc-phase: plan
+agents: [planner]
+skills: [plan-writing, brainstorming, plan-validation]
+commit-types: [docs]
+---
+
+# /plan — Implementation Planning
+
+> **Trigger**: `/plan [task description]`
+> **Lifecycle**: Plan — first step of SDLC after discovery
+
+> Standards: See `rules/workflow-standards.md`
+
+> [!IMPORTANT]
+> This workflow creates plans, NOT code. No implementation during planning. All plans require user approval before execution.
+
+---
+
+## Critical Rules
+
+1. **No code writing** — plans only
+2. **Socratic gate** — ask at least 3 clarifying questions before planning
+3. **Dynamic naming** — `PLAN-{task-slug}.md`
+4. **Verification criteria** — every task must have clear "done" criteria
+5. **User approval required** — never implement without explicit approval
+
+---
+
+## Steps
+
+// turbo
+1. **Clarify Requirements** — ask 3+ clarifying questions about purpose, scope, constraints. Confirm acceptance criteria and edge cases.
+
+// turbo
+2. **Explore Codebase** — scan structure, identify relevant files, patterns, dependencies.
+
+3. **Create Plan**
+   - Consult mandatory rules (security, testing, coding-style, documentation, git-workflow)
+   - Classify task: Trivial (1-2 files), Medium (3-10), Large (10+)
+   - Break down into steps with exact file paths and verification criteria
+   - Include cross-cutting concerns (security, testing, docs) for ALL sizes
+   - For Medium/Large: invoke specialist synthesis (security-reviewer, tdd-guide, architect)
+   - Save to `docs/PLAN-{task-slug}.md`
+
+// turbo
+3.5. **Validate Plan** — verify schema compliance, cross-cutting sections populated, file paths in every step. Score >= 70% → present. Below → revise (max 2 cycles).
+
+4. **Present for Approval** — show plan with quality score, wait for approval.
+
+---
+
+## Output Template
+
+```markdown
+## Plan: [Task Name]
+
+### Scope
+[Coverage and exclusions]
+
+### Tasks
+1. [ ] [Task] — **Verify**: [done criteria]
+
+### Agent Assignments (if multi-domain)
+| Task | Agent | Domain |
+
+### Risks & Considerations
+
+Plan saved: `docs/PLAN-{slug}.md`
+Approve to start with `/create` or `/enhance`.
+```
+
+---
+
+## Governance
+
+**PROHIBITED:** Writing code during planning · proceeding without approval · vague tasks · skipping Socratic gate
+
+**REQUIRED:** 3+ clarifying questions · mandatory rule consultation · verification criteria per task · cross-cutting concerns · plan validation · user approval · plan saved in `docs/`
+
+---
+
+## Post-Implementation Retrospective
+
+After implementation reaches VERIFY phase, compare plan against `git diff --name-only`. Run plan-retrospective protocol, append to `.agent/contexts/plan-quality-log.md`. Non-blocking.
+
+---
+
+## Completion Criteria
+
+- [ ] Requirements clarified, codebase explored
+- [ ] Plan created with verifiable tasks and file paths
+- [ ] Plan validated (score >= 70%)
+- [ ] User approved
+
+---
+
+## Related Resources
+
+- **Previous**: `/brainstorm` · `/quality-gate`
+- **Next**: `/create` · `/enhance`
+- **Skill**: `.agent/skills/plan-writing/SKILL.md`
+- **Agent**: `planner`