forgedev 1.2.0 → 1.3.0

package/docs/prompt-library.md

@@ -1,160 +0,0 @@
-# DevForge Prompt Library
-
-8 workflow guides for developing DevForge. Each workflow includes the exact prompts to use.
-
----
-
-## Flow 1: Add a New Stack
-
-When you want to add support for a new tech stack (e.g., Hono, React+Vite, Express).
-
-**Step 1: Plan**
-```
-I want to add [stack] support to DevForge. Enter plan mode. Research:
-1. What files/directories are needed for a typical [stack] project
-2. What dependencies go in package.json / requirements.txt
-3. What the recommender decision tree should look like
-Write a plan to docs/plans/add-[stack].md
-```
-
-**Step 2: Templates**
-```
-Following the plan in docs/plans/add-[stack].md, create template files in
-templates/[category]/[stack]/. Use {{VARIABLE_NAME}} for substitution.
-Follow the patterns in existing templates like templates/frontend/nextjs/.
-```
-
-**Step 3: Recommender**
-```
-Update src/recommender.js to route to the new [stack] templates.
-Add the new templateModules paths. Update formatStackSummary.
-```
-
-**Step 4: Test**
-```
-Add tests in tests/recommender.test.js for the new stack routing.
-Run npx vitest run to verify all tests pass.
-```
-
----
-
-## Flow 2: Create a New Template
-
-When adding individual template files to an existing stack.
-
-```
-I want to add a [template] to the [stack] stack. Create the template file at
-templates/[category]/[stack]/[path]. Use {{VARIABLE_NAME}} placeholders where
-the project name, description, or config values should go. Check
-src/composer.js buildVariables() for available variables.
-```
-
----
-
-## Flow 3: Fix a Bug
-
-**Step 1: Reproduce**
-```
-There's a bug: [describe]. Write a failing test in tests/ that reproduces it.
-Run npx vitest run to confirm the test fails.
-```
-
-**Step 2: Fix**
-```
-Fix the bug that causes [test name] to fail. Run npx vitest run to confirm
-the fix and that no other tests break.
-```
-
----
-
-## Flow 4: Refactor
-
-```
-I want to refactor [module/function]. Enter plan mode. First:
-1. Check test coverage for the code being refactored
-2. Add tests for any uncovered behavior
-3. Plan the refactoring steps (each should keep tests green)
-Write a plan to docs/plans/refactor-[module].md
-```
-
----
-
-## Flow 5: Add a Feature
-
-For features that touch multiple modules (prompts, recommender, composer, configurator).
-
-**Step 1: Plan**
-```
-I want to add [feature] to DevForge. Enter plan mode. Trace through:
-- src/prompts.js — does the user need to be asked anything new?
-- src/recommender.js — does the decision tree change?
-- src/composer.js — are new template variables needed?
-- src/claude-configurator.js — does the generated infrastructure change?
-- src/uat-generator.js — do UAT scenarios need updating?
-Write a plan to docs/plans/[feature].md
-```
-
-**Step 2: Implement**
-```
-Following docs/plans/[feature].md, implement the feature. Work module by module.
-Run npx vitest run after each module change.
-```
-
----
-
-## Flow 6: Verification
-
-```
-Run /project:verify-all
-```
-
-This launches all 5 agents (code-quality, security, spec-validator, production-readiness, uat-validator) and runs tests.
-
----
-
-## Flow 7: Pre-PR
-
-```
-Run /project:pre-pr
-```
-
-This runs tests, smoke test, code quality review, security review, and checks for staged secrets.
-
----
-
-## Flow 8: UAT
-
-```
-Run /project:run-uat
-```
-
-This reads docs/uat/UAT_TEMPLATE.md, maps scenarios to tests, runs them, and updates UAT_CHECKLIST.csv.
-
----
-
-## Utility Prompts
-
-### "I'm lost"
-```
-Read CLAUDE.md, git log --oneline -10, and git status. Tell me where I am,
-what I was working on, and what I should do next.
-```
-
-### "Is this right?"
-```
-Review my changes (git diff). Check if they follow DevForge patterns:
-ESM imports with .js extensions, chalk for output, path.join for paths,
-{{VARIABLE}} for templates. Flag anything that looks wrong.
-```
-
-### "Before I PR"
-```
-Run /project:pre-pr
-```
-
-### "Explain this code"
-```
-Read [file] and explain what it does, how it fits into the DevForge pipeline
-(prompts → recommender → composer → configurator → uat-generator), and what
-calls it.
-```

package/docs/uat/UAT_CHECKLIST.csv

@@ -1,9 +0,0 @@
-UAT_ID,Scenario,Priority,Automated,Test_File,Status,Last_Run,Notes
-UAT-001,Scaffold Next.js Full-Stack,P0,PARTIAL,tests/composer.test.js,NOT RUN,,Manual smoke test required
-UAT-002,Scaffold FastAPI Backend,P0,PARTIAL,tests/composer.test.js,NOT RUN,,Manual smoke test required
-UAT-003,Scaffold Polyglot Full-Stack,P0,PARTIAL,tests/composer.test.js,NOT RUN,,Manual smoke test required
-UAT-004,Recommender Selects Correct Stack,P0,YES,tests/recommender.test.js,NOT RUN,,
-UAT-005,Template Variable Substitution,P0,YES,tests/composer.test.js,NOT RUN,,
-UAT-006,Claude Code Infrastructure Generated,P1,YES,tests/claude-configurator.test.js,NOT RUN,,
-UAT-007,Invalid Input Handling,P1,NO,,NOT RUN,,Manual test required
-UAT-008,Unsupported Stack Selection,P1,PARTIAL,tests/recommender.test.js,NOT RUN,,

package/docs/uat/UAT_TEMPLATE.md

@@ -1,163 +0,0 @@
-# UAT Scenario Pack: DevForge
-
-## Pre-Conditions
-- [ ] Node.js >= 18 installed
-- [ ] npm available
-- [ ] DevForge dependencies installed (`npm install`)
-- [ ] No existing `test-output/` directory
-
-## Scenarios
-
-### UAT-001: Scaffold Next.js Full-Stack Project — Happy Path
-**Priority:** P0
-**Preconditions:** Clean environment, no test-output/ directory
-**Steps:**
-1. Run `node bin/devforge.js test-output`
-2. Select "Full-stack app"
-3. Select "TypeScript" for language
-4. Select "Yes" for authentication
-5. Select "No" for AI integration
-6. Select "Docker" for deployment
-7. Confirm the recommended stack
-**Expected Result:**
-- `test-output/` directory created
-- Contains `package.json` with Next.js, React, TypeScript, Tailwind, Prisma, NextAuth dependencies
-- Contains `src/app/layout.tsx`, `src/app/page.tsx`
-- Contains `src/app/api/health/route.ts` (health check endpoint)
-- Contains `prisma/schema.prisma`
-- Contains `.claude/` directory with hooks, agents, commands
-- Contains `CLAUDE.md` with Next.js-specific rules
-- Contains `docs/uat/UAT_TEMPLATE.md`
-**Actual Result:** ___
-**Status:** NOT RUN
-**Tester:** ___
-**Date:** ___
-**Notes:** ___
-
-### UAT-002: Scaffold FastAPI Backend Project — Happy Path
-**Priority:** P0
-**Preconditions:** Clean environment, no test-output/ directory
-**Steps:**
-1. Run `node bin/devforge.js test-output`
-2. Select "API / backend service"
-3. Select "Python" for language
-4. Select "Yes" for authentication
-5. Select "No" for AI integration
-6. Select "Docker" for deployment
-7. Confirm the recommended stack
-**Expected Result:**
-- `test-output/` directory created
-- Contains `backend/requirements.txt` with FastAPI, SQLAlchemy, Pydantic
-- Contains `backend/app/main.py` with health endpoint and graceful shutdown
-- Contains `backend/app/api/health.py`
-- Contains `backend/app/core/config.py`, `errors.py`, `retry.py`
-- Contains `backend/tests/` with pytest fixtures
-- Contains `.claude/` directory with Python-specific hooks
-- Contains `CLAUDE.md` with FastAPI-specific rules
-**Actual Result:** ___
-**Status:** NOT RUN
-**Tester:** ___
-**Date:** ___
-**Notes:** ___
-
-### UAT-003: Scaffold Polyglot Full-Stack Project — Happy Path
-**Priority:** P0
-**Preconditions:** Clean environment, no test-output/ directory
-**Steps:**
-1. Run `node bin/devforge.js test-output`
-2. Select "Full-stack app"
-3. Select "TypeScript" and "Python" for language
-4. Select "Yes" for authentication
-5. Select "Yes" for AI integration
-6. Select "Docker" for deployment
-7. Confirm the recommended stack
-**Expected Result:**
-- `test-output/` directory created
-- Contains both `frontend/` and `backend/` directories
-- Contains `docker-compose.yml` at root
-- Contains Next.js frontend and FastAPI backend
-- Contains both Prisma and SQLAlchemy database configs
-- Contains polyglot Claude Code hooks (TypeScript + Python)
-**Actual Result:** ___
-**Status:** NOT RUN
-**Tester:** ___
-**Date:** ___
-**Notes:** ___
-
-### UAT-004: Recommender Selects Correct Stack
-**Priority:** P0
-**Preconditions:** None
-**Steps:**
-1. Run `npx vitest run tests/recommender.test.js`
-2. Verify all test cases pass
-3. Verify: web_app + TypeScript → Next.js full-stack
-4. Verify: api_service + Python → FastAPI backend
-5. Verify: full_stack + TS + Python → polyglot
-6. Verify: unsupported combos return helpful error
-**Expected Result:** All recommender tests pass, all 3 stacks correctly selected
-**Actual Result:** ___
-**Status:** NOT RUN
-**Tester:** ___
-**Date:** ___
-**Notes:** ___
-
-### UAT-005: Template Variable Substitution
-**Priority:** P0
-**Preconditions:** None
-**Steps:**
-1. Run `npx vitest run tests/composer.test.js`
-2. Verify `{{PROJECT_NAME}}` replaced in all .template files
-3. Verify non-.template files copied without modification
-4. Verify .gitkeep files preserved
-5. Verify no `{{` patterns remain in output
-**Expected Result:** All composer tests pass, variables correctly substituted
-**Actual Result:** ___
-**Status:** NOT RUN
-**Tester:** ___
-**Date:** ___
-**Notes:** ___
-
-### UAT-006: Claude Code Infrastructure Generated
-**Priority:** P1
-**Preconditions:** Scaffold a project first
-**Steps:**
-1. Run `npx vitest run tests/claude-configurator.test.js`
-2. Verify `.claude/settings.json` created with correct hooks
-3. Verify CLAUDE.md generated with stack-specific content
-4. Verify agents copied (5 agents)
-5. Verify skills copied (filtered by stack)
-6. Verify commands copied (6 commands)
-**Expected Result:** All claude-configurator tests pass, infrastructure complete
-**Actual Result:** ___
-**Status:** NOT RUN
-**Tester:** ___
-**Date:** ___
-**Notes:** ___
-
-### UAT-007: Invalid Input Handling
-**Priority:** P1
-**Preconditions:** None
-**Steps:**
-1. Run `node bin/devforge.js` (no project name) — should show error
-2. Run `node bin/devforge.js .` (invalid name) — should show error
-3. Create `test-output/` dir, then run `node bin/devforge.js test-output` — should warn about existing dir
-**Expected Result:** Clear error messages, no crashes, exit code 1
-**Actual Result:** ___
-**Status:** NOT RUN
-**Tester:** ___
-**Date:** ___
-**Notes:** ___
-
-### UAT-008: Unsupported Stack Selection
-**Priority:** P1
-**Preconditions:** None
-**Steps:**
-1. Run `node bin/devforge.js test-output`
-2. Select "Mobile app" or "Desktop app"
-3. Observe the recommendation
-**Expected Result:** Displays message that the stack is not yet supported in V1, suggests closest supported option
-**Actual Result:** ___
-**Status:** NOT RUN
-**Tester:** ___
-**Date:** ___
-**Notes:** ___

package/templates/claude-code/commands/done.md

@@ -1,19 +0,0 @@
-Verify that the current task is actually complete before moving on.
-
-## Checklist
-
-1. Run lint and confirm it passes: `{{LINT_COMMAND}}`
-2. Run type check and confirm it passes: `{{TYPE_CHECK_COMMAND}}`
-3. Run tests and confirm they pass: `{{TEST_COMMAND}}`
-4. Check that no `.env` files or secrets are in the changes
-5. Review the git diff to make sure only intended files changed
-6. Verify the feature works as described (manual check or UAT scenario)
-
-## Output
-
-If everything passes:
-```
-Task complete. Ready to commit or move on.
-```
-
-If something fails, explain what needs to be fixed before the task can be considered done.