opencode-goopspec 0.1.4 → 0.1.6

package/references/xml-response-schema.md

@@ -14,7 +14,7 @@ All GoopSpec agents use a standardized XML envelope for machine-parseable handof
 Every agent response MUST include this XML block at the END of their response:
 
 ```xml
-<goop_report version="0.1.4">
+<goop_report version="0.1.6">
   <status>COMPLETE|PARTIAL|BLOCKED|CHECKPOINT</status>
   <agent>goop-[type]</agent>
   <task_id>W[wave].T[task]</task_id>
@@ -100,7 +100,7 @@ Every agent response MUST include this XML block at the END of their response:
 
 ### Executor Response
 ```xml
-<goop_report version="0.1.4">
+<goop_report version="0.1.6">
   <status>COMPLETE</status>
   <agent>goop-executor</agent>
   <task_id>W2.T3</task_id>
@@ -149,7 +149,7 @@ Every agent response MUST include this XML block at the END of their response:
 
 ### Planner Response
 ```xml
-<goop_report version="0.1.4">
+<goop_report version="0.1.6">
   <status>COMPLETE</status>
   <agent>goop-planner</agent>
   <task_name>Create execution blueprint</task_name>
@@ -188,7 +188,7 @@ Every agent response MUST include this XML block at the END of their response:
 
 ### Blocked Response
 ```xml
-<goop_report version="0.1.4">
+<goop_report version="0.1.6">
   <status>BLOCKED</status>
   <agent>goop-executor</agent>
   <task_id>W2.T2</task_id>
@@ -233,4 +233,4 @@ Every agent response MUST include this XML block at the END of their response:
 
 ---
 
-*GoopSpec XML Response Schema v0.1.4*
+*GoopSpec XML Response Schema v0.1.6*

package/skills/accessibility/skill.md

@@ -7,7 +7,7 @@ triggers:
   - wcag
   - a11y
   - inclusive
-version: 0.1.0
+version: 0.1.6
 ---
 
 # Accessibility Skill

package/skills/accessibility-testing/skill.md

@@ -7,7 +7,7 @@ triggers:
   - wcag
   - a11y
   - testing
-version: 0.1.0
+version: 0.1.6
 ---
 
 # Accessibility Testing Skill

package/skills/api-docs/skill.md

@@ -7,7 +7,7 @@ triggers:
   - documentation
   - reference
   - openapi
-version: 0.1.0
+version: 0.1.6
 ---
 
 # API Documentation Skill

package/skills/architecture-design/skill.md

@@ -7,7 +7,7 @@ triggers:
   - design
   - structure
   - system
-version: 0.1.0
+version: 0.1.6
 requires:
   - goop-core
 ---

package/skills/atomic-commits/skill.md

@@ -7,47 +7,124 @@ triggers:
   - git
   - atomic
   - history
-version: 0.1.0
+version: 0.1.6
 ---
 
 # Atomic Commits Skill
 
 ## Purpose
-Create small, focused commits that each represent a single logical change.
 
-## Principles
+Create small, focused commits that each represent a single logical change. All commits must be professional and universally understandable.
+
+## Core Principles
 
 1. **One Change Per Commit** - Each commit should do one thing
 2. **Buildable State** - Every commit should leave the code in a working state
 3. **Meaningful Messages** - Commit messages explain the why, not just the what
 4. **Reviewable Size** - Commits should be easy to review (< 200 lines ideal)
+5. **Universal Language** - No internal jargon, phases, or task references
+
+## Critical: Universal Commit Messages
+
+**NEVER include in commit messages:**
+- GoopSpec phases, waves, or task IDs (W1.T2, Phase 3, etc.)
+- References to SPEC.md, BLUEPRINT.md, or planning documents
+- Agent names or orchestration terminology
+- Internal must-have IDs (MH-01, etc.)
+
+Write commits as if no one on the team knows GoopSpec exists.
 
 ## Commit Structure
 
 ```
-type(scope): short description
+type(scope): concise but descriptive title (max 72 chars)
 
-Longer explanation of why this change was made.
-Include context that helps future readers understand
-the decision.
+[2-4 sentence paragraph explaining context and motivation.
+Why was this change needed? What problem does it solve?]
 
-Closes #123
+Changes:
+- Specific change with context
+- Another change with why it matters
+
+[Optional: Breaking changes, migration notes]
 ```
 
 ## Types
 
-- `feat` - New feature
-- `fix` - Bug fix
-- `refactor` - Code restructuring
-- `docs` - Documentation only
-- `test` - Test additions/changes
-- `chore` - Maintenance tasks
-- `perf` - Performance improvements
+| Type | Use For |
+|------|---------|
+| `feat` | New feature or capability |
+| `fix` | Bug fix |
+| `refactor` | Code restructuring (no behavior change) |
+| `docs` | Documentation only |
+| `test` | Adding or updating tests |
+| `chore` | Config, deps, build tooling |
+| `style` | Formatting, whitespace |
+| `perf` | Performance improvements |
+
+## Good vs Bad Examples
+
+**Good:**
+```
+feat(auth): Add OAuth2 login with Google and GitHub
+
+Users can now sign in using their Google or GitHub accounts.
+This reduces friction for new user onboarding and leverages
+existing identity providers for security.
+
+Changes:
+- Add OAuth2 client configuration
+- Implement callback handlers for each provider
+- Add provider selection UI to login page
+```
+
+**Bad:**
+```
+feat(auth): Wave 2 Task 4 - OAuth implementation
+
+Completed W2.T4 from BLUEPRINT.md per MH-05 specification.
+goop-executor implemented this feature.
+
+Changes:
+- Implemented as specified in SPEC.md
+```
+
+## Single vs Multiple Commits
+
+**Single commit when:**
+- All changes serve one purpose
+- Changes are tightly coupled
+- Small scope (<100 lines, <5 files)
+
+**Multiple commits when:**
+- Unrelated fixes mixed with features
+- Refactoring separate from new functionality
+- Tests added independently
+- Config changes separate from code
+
+## Pre-Commit Checklist
+
+- [ ] Tests pass
+- [ ] No TypeScript/linting errors
+- [ ] No console.log or debug code
+- [ ] Message is specific (not generic)
+- [ ] Message explains WHY
+- [ ] No internal references (phases, waves, tasks)
+- [ ] GPG signing preserved
+
+## Reference
+
+For complete Git workflow guidance, load:
+```
+goop_reference({ name: "git-workflow" })
+```
 
 ## Anti-Patterns to Avoid
 
 - "WIP" commits
 - "Fix typo" without context
+- "Update files" or "Minor changes"
 - Multiple unrelated changes in one commit
 - Breaking commits (code doesn't build)
 - Giant commits that do everything
+- Internal jargon in messages

package/skills/code-review/skill.md

@@ -7,7 +7,7 @@ triggers:
   - feedback
   - pull request
   - quality
-version: 0.1.0
+version: 0.1.6
 ---
 
 # Code Review Skill

package/skills/codebase-mapping/skill.md

@@ -7,7 +7,7 @@ triggers:
   - architecture
   - discovery
   - mapping
-version: 0.1.0
+version: 0.1.6
 ---
 
 # Codebase Mapping Skill

package/skills/convention-detection/skill.md

@@ -7,7 +7,7 @@ triggers:
   - style
   - patterns
   - guidelines
-version: 0.1.0
+version: 0.1.6
 ---
 
 # Convention Detection Skill

package/skills/debugging/skill.md

@@ -7,7 +7,7 @@ triggers:
   - bug
   - diagnose
   - fix
-version: 0.1.0
+version: 0.1.6
 ---
 
 # Debugging Skill

package/skills/deviation-handling/skill.md

@@ -7,7 +7,7 @@ triggers:
   - unexpected
   - bug
   - blocker
-version: 0.1.0
+version: 0.1.6
 requires:
   - goop-core
 ---

package/skills/documentation/skill.md

@@ -7,7 +7,7 @@ triggers:
   - readme
   - api
   - guide
-version: 0.1.0
+version: 0.1.6
 requires:
   - goop-core
 ---

package/skills/goop-core/skill.md

@@ -1,6 +1,6 @@
 ---
 name: goop-core
-description: Core GoopSpec 0.1.0 operations - the 5-phase spec-driven workflow
+description: Core GoopSpec 0.1.6 operations - the 5-phase spec-driven workflow
 category: core
 triggers:
   - goop
@@ -8,10 +8,10 @@ triggers:
   - workflow
   - plan
   - execute
-version: 0.1.0
+version: 0.1.6
 ---
 
-# GoopSpec 0.1.0 Core Operations
+# GoopSpec 0.1.6 Core Operations
 
 ## The 5-Phase Workflow
 
@@ -299,25 +299,62 @@ STOP and ask user for:
 
 ---
 
-## Commit Format
+## Git Workflow
+
+**CRITICAL:** All commits and PRs must use universal language. Never reference:
+- GoopSpec phases, waves, or task IDs (W1.T2, Phase 3, etc.)
+- Internal documents (SPEC.md, BLUEPRINT.md, must-have IDs)
+- Agent names or orchestration concepts
+
+Write as if no one knows GoopSpec exists.
+
+### Commit Format
 
 ```
-type(wave-task): description
+type(scope): concise but descriptive title (max 72 chars)
+
+[2-4 sentence paragraph explaining context and motivation.
+Why was this change needed? What problem does it solve?]
 
-- Detail 1
-- Detail 2
+Changes:
+- Specific change with context
+- Another change with why it matters
 ```
 
 Types: `feat`, `fix`, `test`, `refactor`, `docs`, `perf`, `chore`
 
-Example:
+**Good Example:**
 ```
-feat(w1-t2): implement user authentication API
+feat(auth): Add OAuth2 login with Google and GitHub
 
+Users can now sign in using their Google or GitHub accounts.
+This reduces friction for new user onboarding and improves security.
+
+Changes:
 - Add /auth/login and /auth/logout endpoints
-- Integrate JWT token generation
-- Add refresh token rotation
+- Integrate JWT token generation with OAuth providers
+- Add refresh token rotation for session persistence
+```
+
+**Bad Example:**
 ```
+feat(w1-t2): implement user authentication API (per MH-03)
+```
+
+### Branch Naming
+
+At session start, agents offer to create a feature branch:
+- Format: `type/short-description`
+- Types: `feat/`, `fix/`, `refactor/`, `chore/`
+
+### Pull Requests
+
+At end of phases, agents offer to create PRs:
+- Title: `type(scope): Descriptive summary`
+- Body: Summary, Changes, Testing, Notes
+- No internal terminology
+
+For complete guidance: `goop_reference({ name: "git-workflow" })`
 
 ---
 

package/skills/memory-usage/skill.md

@@ -8,7 +8,7 @@ triggers:
   - recall
   - save for later
   - persistent
-version: 0.1.0
+version: 0.1.6
 ---
 
 # Memory System Usage Guide

package/skills/parallel-planning/skill.md

@@ -7,7 +7,7 @@ triggers:
   - wave
   - concurrent
   - batch
-version: 0.1.0
+version: 0.1.6
 requires:
   - goop-core
   - task-decomposition

package/skills/pattern-extraction/skill.md

@@ -7,7 +7,7 @@ triggers:
   - abstraction
   - reuse
   - refactor
-version: 0.1.0
+version: 0.1.6
 ---
 
 # Pattern Extraction Skill

package/skills/performance-optimization/skill.md

@@ -7,7 +7,7 @@ triggers:
   - optimize
   - speed
   - memory
-version: 0.1.0
+version: 0.1.6
 requires:
   - goop-core
 ---

package/skills/playwright/skill.md

@@ -7,7 +7,7 @@ triggers:
   - e2e
   - browser
   - testing
-version: 0.1.0
+version: 0.1.6
 ---
 
 # Playwright Skill

package/skills/playwright-testing/skill.md

@@ -7,7 +7,7 @@ triggers:
   - testing
   - e2e
   - suites
-version: 0.1.0
+version: 0.1.6
 ---
 
 # Playwright Testing Skill

package/skills/progress-tracking/skill.md

@@ -7,7 +7,7 @@ triggers:
   - status
   - track
   - todo
-version: 0.1.0
+version: 0.1.6
 requires:
   - goop-core
 ---

package/skills/readme-generation/skill.md

@@ -7,7 +7,7 @@ triggers:
   - documentation
   - onboarding
   - usage
-version: 0.1.0
+version: 0.1.6
 ---
 
 # README Generation Skill

package/skills/research/skill.md

@@ -7,7 +7,7 @@ triggers:
   - investigate
   - explore
   - learn
-version: 0.1.0
+version: 0.1.6
 requires:
   - goop-core
 ---

package/skills/responsive-design/skill.md

@@ -7,7 +7,7 @@ triggers:
   - mobile
   - breakpoints
   - layout
-version: 0.1.0
+version: 0.1.6
 ---
 
 # Responsive Design Skill

package/skills/scientific-method/skill.md

@@ -7,7 +7,7 @@ triggers:
   - experiment
   - evidence
   - method
-version: 0.1.0
+version: 0.1.6
 ---
 
 # Scientific Method Skill

package/skills/security-audit/skill.md

@@ -7,7 +7,7 @@ triggers:
   - vulnerability
   - audit
   - pentest
-version: 0.1.0
+version: 0.1.6
 requires:
   - goop-core
 ---

package/skills/task-decomposition/skill.md

@@ -7,7 +7,7 @@ triggers:
   - breakdown
   - split
   - tasks
-version: 0.1.0
+version: 0.1.6
 requires:
   - goop-core
 ---

package/skills/task-delegation/skill.md

@@ -7,7 +7,7 @@ triggers:
   - spawn
   - subagent
   - task
-version: 0.1.0
+version: 0.1.6
 requires:
   - goop-core
 ---
@@ -94,31 +94,49 @@ Don't pass:
 - Unrelated file contents
 - Completed task details
 
-## The Dispatch Tool (CRITICAL)
+## Two-Step Delegation (CRITICAL)
 
-**ALWAYS use the native `task` tool to dispatch agents.**
+Delegation in GoopSpec uses **two distinct tools** with different purposes:
+
+### Step 1: `goop_delegate` — Prompt Engineering
+
+This tool **prepares** a rich, production-ready prompt by:
+- Loading the agent's definition (skills, references, model)
+- Injecting team awareness context
+- Adding memory protocols
+- Checking for file conflicts with other agents
+
+```typescript
+goop_delegate({
+  agent: "goop-executor",
+  prompt: "Implement user authentication",
+  context: "Stack: Next.js + NextAuth"
+})
+```
+
+**Output**: A structured prompt package with the exact `task()` call to execute.
+
+### Step 2: `task` — Agent Execution
+
+This tool **spawns** the subagent with the engineered prompt:
 
 ```typescript
 task({
-  subagent_type: "goop-executor",  // Use goop-[agent-name]
-  description: "Implement authentication",
-  prompt: `
-    ## TASK
-    Implement user authentication
-
-    ## CONTEXT
-    - SPEC: .goopspec/SPEC.md
-    - BLUEPRINT: .goopspec/BLUEPRINT.md
-    
-    ## REQUIREMENTS
-    [Specific requirements from SPEC.md]
-    
-    ## VERIFICATION
-    [How to confirm task completion]
-  `
+  subagent_type: "goop-executor",
+  description: "Implement auth",
+  prompt: `[Copy composedPrompt from goop_delegate output]`
 })
 ```
 
+### When to Use Each Pattern
+
+| Situation | Recommended Pattern |
+|-----------|---------------------|
+| Complex task needing skills/refs | `goop_delegate` → `task` |
+| Task needing team awareness | `goop_delegate` → `task` |
+| Simple, well-defined task | `task` directly |
+| Quick exploration | `task` directly |
+
 ### Available subagent_types
 
 | subagent_type | Use For |
@@ -135,28 +153,36 @@ task({
 | `goop-librarian` | Code/docs search, information retrieval |
 | `general` | Fallback for any task |
 
-### Do NOT Use These Tools for Delegation
+### Common Mistakes
 
-| Tool | Why NOT |
-|------|---------|
-| `delegate` | Different system (async delegation), NOT GoopSpec agents |
-| `goop_delegate` alone | Only composes prompts, doesn't execute - must follow with `task` |
+| Mistake | Problem | Fix |
+|---------|---------|-----|
+| Using `goop_delegate` without `task` | Prompt is prepared but agent never runs | Always follow `goop_delegate` with `task` |
+| Using `delegate` instead | Different system, not GoopSpec agents | Use `task` for GoopSpec agents |
+| Skipping `goop_delegate` for complex tasks | Miss skills/refs/team context | Use full two-step for complex work |
 
-### Optional: goop_delegate for Rich Prompts
-
-Use `goop_delegate` only when you need skills/references auto-injected:
+### Full Example
 
 ```typescript
-// Step 1: Compose prompt with goop_delegate (optional)
-const result = goop_delegate({ agent: "goop-executor", prompt: "..." })
-// Returns: <goop_delegation> JSON with composedPrompt
+// Step 1: Engineer the prompt
+goop_delegate({
+  agent: "goop-executor",
+  prompt: "Implement password reset flow",
+  context: `
+    Stack: Next.js 14, Prisma, Resend
+    Task from BLUEPRINT.md Wave 2, Task 3
+    Must follow existing auth patterns in src/auth/
+  `
+})
 
-// Step 2: Execute with task (REQUIRED)
-task({ subagent_type: "goop-executor", prompt: composedPrompt })
+// Step 2: Execute (copy the task call from goop_delegate output)
+task({
+  subagent_type: "goop-executor",
+  description: "Implement password reset",
+  prompt: `[The full composedPrompt from goop_delegate]`
+})
 ```
 
-**For most cases: Just use `task` directly.**
-
 ## Error Handling
 
 If delegated task fails:

package/skills/technical-writing/skill.md

@@ -7,7 +7,7 @@ triggers:
   - documentation
   - docs
   - guide
-version: 0.1.0
+version: 0.1.6
 ---
 
 # Technical Writing Skill

package/skills/testing/skill.md

@@ -7,7 +7,7 @@ triggers:
   - spec
   - assert
   - coverage
-version: 0.1.0
+version: 0.1.6
 requires:
   - goop-core
 ---

package/skills/ui-design/skill.md

@@ -7,7 +7,7 @@ triggers:
   - visual
   - layout
   - interface
-version: 0.1.0
+version: 0.1.6
 ---
 
 # UI Design Skill

package/skills/ux-patterns/skill.md

@@ -7,7 +7,7 @@ triggers:
   - patterns
   - usability
   - interaction
-version: 0.1.0
+version: 0.1.6
 ---
 
 # UX Patterns Skill