@comfanion/workflow 4.36.60 → 4.36.62

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@comfanion/workflow",
-  "version": "4.36.60",
+  "version": "4.36.62",
   "description": "Initialize OpenCode Workflow system for AI-assisted development with semantic code search",
   "type": "module",
   "bin": {

package/src/build-info.json

@@ -1,6 +1,6 @@
 {
-  "version": "4.36.60",
-  "buildDate": "2026-01-26T01:30:41.983Z",
+  "version": "4.36.62",
+  "buildDate": "2026-01-26T02:24:41.062Z",
   "files": [
     "config.yaml",
     "FLOW.yaml",

package/src/opencode/agents/analyst.md

@@ -17,7 +17,7 @@ tools:
   list: true
   skill: true
   question: true
-  bash: false        # No shell commands needed
+  bash: true         # For file operations
   webfetch: true     # For research
   todowrite: true    # Track complex requirements gathering
   todoread: true
@@ -27,7 +27,13 @@ tools:
 # Permissions - granular control
 permission:
   edit: allow        # Can write documentation
-  bash: deny         # No bash access
+  bash:
+    "*": ask
+    "ls *": allow
+    "cat *": allow
+    "tree *": allow
+    "mkdir *": allow
+    "mv *": allow
 ---
 
 <agent id="analyst" name="Sara" title="Business Analyst" icon="📊">

package/src/opencode/agents/architect.md

@@ -18,7 +18,7 @@ tools:
   skill: true
   question: true
   bash: true         # For codebase analysis
-  webfetch: false    # Use @researcher for web research
+  webfetch: true     # For technical research
   todowrite: true    # Architecture can be complex multi-step
   todoread: true
   lsp: true          # Code intelligence for architecture analysis
@@ -39,6 +39,7 @@ permission:
     "mkdir *": allow
     "head *": allow
     "tail *": allow
+    "mv *": allow
 ---
 
 <agent id="architect" name="Winston" title="Solution Architect" icon="🏗️">

package/src/opencode/agents/coder.md

@@ -56,14 +56,6 @@ permission:
   </step>
   <step n="8">Report completion or errors</step>
 
-  <lint-commands hint="Common linter commands">
-    <js>npx eslint --fix {files} OR npx biome check --write {files}</js>
-    <ts>npx eslint --fix {files} OR npx tsc --noEmit</ts>
-    <go>gofmt -w {files} && golangci-lint run {files}</go>
-    <py>ruff check --fix {files} OR black {files}</py>
-    <rust>cargo fmt && cargo clippy --fix</rust>
-  </lint-commands>
-
   <rules>
     <r>DO NOT ask clarifying questions - execute or fail</r>
     <r>DO NOT refactor beyond task scope</r>

package/src/opencode/agents/pm.md

@@ -20,7 +20,7 @@ tools:
   skill: true
   question: true
   bash: true         # Limited bash for directory ops
-  webfetch: false    # Use @researcher for web research
+  webfetch: true     # For product research
   todowrite: true    # PM tracks complex planning tasks
   todoread: true
   search: true       # PREFERRED: Semantic search for docs/PRD/architecture
@@ -35,6 +35,7 @@ permission:
     "cat *": allow
     "tree *": allow
     "mkdir *": allow
+    "mv *": allow
     "git branch*": allow
     "git status": allow
     "git log*": allow

package/src/opencode/commands/archive.md

@@ -5,23 +5,6 @@ agent: pm
 
 # Archive Command
 
-## Arguments
-$ARGUMENTS
-
-Format: `[action] [path]`
-
-**Actions:**
-- "move [path]" - Archive a specific document
-- "sprint [sprint-N]" - Archive completed sprint
-- "list" - Show archive contents
-- "find [query]" - Search archive
-
-**Examples:**
-- `/archive move docs/research/technical/old-topic.md`
-- `/archive sprint sprint-0`
-- `/archive list`
-- `/archive find authentication`
-
 ## Archive Policy
 
 **CRITICAL RULES:**

package/src/opencode/commands/code-review.md

@@ -7,20 +7,6 @@ agent: dev
 
 Review implemented code for quality, correctness, and adherence to standards.
 
-## Usage
-
-```
-/code-review [story-path]
-```
-
-## Arguments
-
-- `story-path` (optional): Path to story file to review. If not provided, finds stories in `review` status.
-
-## Agent
-
-This command invokes the **Dev** agent (Rick) in review mode.
-
 ## Process
 
 1. Load story file and implementation
@@ -78,16 +64,6 @@ Updates story file with:
 - [ ] No SQL injection risks
 - [ ] Proper authentication/authorization
 
-## Example
-
-```bash
-# Review stories in 'review' status
-/code-review
-
-# Review specific story
-/code-review docs/sprint-artifacts/sprint-1/stories/1-1-user-auth.md
-```
-
 ## After Review
 
 - If **Approved**: Mark story as `done`

package/src/opencode/commands/dev-story.md

@@ -7,20 +7,6 @@ agent: dev
 
 Implement a story using red-green-refactor cycle with TODO tracking.
 
-## Usage
-
-```
-/dev-story [story-path]
-```
-
-## Arguments
-
-- `story-path` (optional): Path to specific story file. If not provided, finds next `ready-for-dev` story from sprint-status.yaml.
-
-## Agent
-
-This command invokes the **Dev** agent (Rick).
-
 ## Process
 
 ### Phase 1: Setup
@@ -100,16 +86,6 @@ The workflow will HALT and ask for input when:
 - Required configuration is missing
 - Ambiguous requirements need clarification
 
-## Example
-
-```bash
-# Find next ready story automatically
-/dev-story
-
-# Implement specific story
-/dev-story docs/sprint-artifacts/sprint-1/stories/1-1-user-auth.md
-```
-
 ## Story Status Flow
 
 ```

package/src/opencode/commands/diagram.md

@@ -5,32 +5,6 @@ agent: architect
 
 # Diagram Creation
 
-## Arguments
-$ARGUMENTS
-
-Format: `[scope] [type]`
-
-**Scope:**
-- "system" - System-wide diagram
-- "[module-name]" - Module-specific diagram
-
-**Type:**
-- "c4-context" - C4 Level 1: System context
-- "c4-container" - C4 Level 2: Containers/services
-- "c4-component" - C4 Level 3: Module components
-- "sequence [flow-name]" - Sequence diagram for a flow
-- "erd" - Entity relationship diagram
-- "flow [process-name]" - Flowchart
-- "state [entity-name]" - State machine
-
-**Examples:**
-- `/diagram system c4-context`
-- `/diagram catalog c4-component`
-- `/diagram catalog sequence product-import`
-- `/diagram catalog erd`
-- `/diagram order flow checkout-process`
-- `/diagram offer state lifecycle`
-
 ## Check Existing Diagrams
 
 !`ls -la docs/diagrams/ 2>/dev/null || echo "No diagrams yet"`

package/src/opencode/commands/quick.md

@@ -13,18 +13,6 @@ Skip the full workflow for small changes (L0-L1 scale):
 - Configuration changes
 - Small refactors
 
-## Arguments
-$ARGUMENTS
-
-Description of the change to make.
-
-Example:
-```
-/quick Fix login button not working on mobile
-/quick Add dark mode toggle to settings
-/quick Update API endpoint for user service
-```
-
 ## Quick Mode Criteria
 
 Automatically used when:

package/src/opencode/commands/research.md

@@ -5,29 +5,6 @@ agent: researcher
 
 # Research Command
 
-## Arguments
-$ARGUMENTS
-
-Format: `[type] [topic]`
-
-**Types:**
-- "technical" - Technology evaluation
-- "market" - Market/competitor analysis
-- "domain" - Business domain research
-- "integration" - External system/API research
-- "pattern" - Architecture/design patterns
-
-**Examples:**
-- `/research technical message-queues`
-- `/research market competitor-analysis`
-- `/research domain e-commerce-regulations`
-- `/research integration stripe-api`
-- `/research pattern cqrs-event-sourcing`
-
-**Other commands:**
-- `/research list` - Show all research
-- `/research update [path]` - Update existing research
-
 ## Check Existing Research
 
 !`tree docs/research/ 2>/dev/null || echo "No research yet"`

package/src/opencode/commands/review-story.md

@@ -7,20 +7,6 @@ agent: reviewer
 
 Review a completed story using @reviewer agent (GPT-5.2 Codex) for deep security and quality analysis.
 
-## Usage
-
-```
-/review-story [story-path]
-```
-
-## Arguments
-
-- `story-path` (optional): Path to story file. If not provided, finds stories in `review` status.
-
-## Agent
-
-This command invokes the **Reviewer** agent (Marcus) with GPT-5.2 Codex model - best at finding bugs and security issues.
-
 ## When to Use
 
 1. **After `/dev-story`** completes all tasks (auto-invoked if `auto_review: true`)
@@ -104,16 +90,6 @@ development:
   auto_review: true   # Auto-invoke @reviewer after /dev-story completes
 ```
 
-## Example
-
-```bash
-# Review stories in 'review' status
-/review-story
-
-# Review specific story
-/review-story docs/sprint-artifacts/sprint-1/stories/story-01-user-auth.md
-```
-
 ## Flow with /dev-story
 
 ```

package/src/opencode/commands/search.md

@@ -7,17 +7,6 @@ agent: crawler
 
 Semantic search across the codebase using vectorizer.
 
-## Arguments
-$ARGUMENTS
-
-## Examples
-
-```
-/search authentication middleware
-/search how users are stored
-/search error handling patterns
-```
-
 ## Your Task
 
 Perform **multiple semantic searches** to find the best results for user's query.

package/src/opencode/commands/unit-docs.md

@@ -9,14 +9,6 @@ skills_loaded:
 
 Create structured documentation for any logical piece of the system using the Universal Unit format.
 
-## Arguments
-
-$ARGUMENTS
-
-- `[unit-type] [unit-name]`: Create unit doc (e.g., `entity Task`, `module catalog`)
-- `list`: List existing unit docs
-- `validate [unit-name]`: Validate unit doc completeness
-
 ## Supported Unit Types
 
 | Type | When to Use | Example |
@@ -150,21 +142,4 @@ Suggest:
 - `/unit-docs [type] [name]` - Document related units
 - `/stories {epic-id}` - Reference unit in story tasks
 
-## Examples
-
-```bash
-# Document a core entity
-/unit-docs entity Task
 
-# Document a service
-/unit-docs service NotificationService
-
-# Document a module
-/unit-docs module catalog
-
-# List all units
-/unit-docs list
-
-# Validate a unit
-/unit-docs validate Task
-```

package/src/opencode/skills/architecture-validation/SKILL.md

@@ -170,7 +170,7 @@ Orders ──→ Catalog, Inventory
 ## Next Steps
 
 After validation:
-- If PASS, proceed to `/epics`
+- If PASS, proceed to epic creation
 ```
 
 ## If QA Artifact Missing

package/src/opencode/skills/dev-story/SKILL.md

@@ -11,123 +11,112 @@ metadata:
 
 # Dev Story Skill
 
-Details for implementing stories. For workflow overview, see `/dev-story` command.
+How to transform story tasks into executable instructions for @coder.
 
-## Critical Rules
+## Task Transformation
 
-- **Story File is the single source of truth** - tasks sequence is authoritative
-- **Respect task dependencies** - never start blocked tasks
-- **Tests are MANDATORY** - each task has tests that MUST pass
-- **NEVER lie about tests** - tests must actually exist and pass
-- **For parallel execution** - call multiple @coder in one message
+Story task is a specification. @coder needs executable instruction with full context.
 
-## Methodology (from config.yaml)
+### Add Context
 
-| Methodology | Flow | Validation |
-|-------------|------|------------|
-| **TDD** | Interface → Test (RED) → Impl (GREEN) → Refactor | Test must FAIL first, then PASS |
-| **STUB** | Interface → Stub → Test → Real Impl | Test against stub, then real |
+@coder doesn't see the full story. Include:
+- **Existing files** - actual paths to domain entities, interfaces, services that exist
+- **Patterns to follow** - link to existing similar code ("follow pattern from X")
+- **What was done** - results of previous tasks that this task depends on
+- **Imports** - what packages to use
 
-## Task Implementation
+### Add Implementation Details
 
-### Dependency Check
+Story has Code Sketch. Expand it:
+- Actual method signatures with types
+- Error handling approach
+- Logging requirements
+- Validation rules
 
-```xml
-<phase name="DEPENDENCY_CHECK">
-  <action>Parse task summary table from story file</action>
-  <action>Read "Depends On" column for current task</action>
-  <check if="dependencies not complete">
-    <action>SKIP this task, find next ready task</action>
-  </check>
-</phase>
-```
+### Add Verification
 
-### Parallel Opportunity
+Story has "Done when". Make it concrete:
+- Specific test commands to run
+- Expected output format
+- Files that must compile
 
-```xml
-<phase name="PARALLEL_CHECK">
-  <action>Identify tasks with same satisfied dependencies</action>
-  <check if="parallel tasks exist">
-    <action>Call multiple @coder in ONE message</action>
-  </check>
-</phase>
-```
+## Formulating Task for @coder
 
-### TDD Cycle
-
-```xml
-<check if="methodology is TDD">
-  <phase name="RED">
-    <action>Write test for task deliverables</action>
-    <action>Run test - MUST FAIL</action>
-  </phase>
-
-  <phase name="GREEN">
-    <action>Implement MINIMAL code to pass test</action>
-    <action>Run test - MUST PASS</action>
-  </phase>
-
-  <phase name="REFACTOR">
-    <action>Improve code structure</action>
-    <action>Run test - MUST STILL PASS</action>
-  </phase>
-</check>
+Structure:
 ```
+## Task: [Name] (Task IDs)
 
-### STUB Cycle
+[One line goal]
 
-```xml
-<check if="methodology is STUB">
-  <phase name="STUB">
-    <action>Write stub implementation (mock data)</action>
-  </phase>
+### Skills
+- Use `doc-todo` for TODO comments
 
-  <phase name="TEST">
-    <action>Write tests against stub</action>
-    <action>Run test - MUST PASS with stub</action>
-  </phase>
+### Context
+- [Existing file]: [What it contains, what to use from it]
+- [Another file]: [Description]
+- Existing pattern: [Path to similar code to follow]
 
-  <phase name="REAL">
-    <action>Replace stub with real implementation</action>
-    <action>Run test - MUST STILL PASS</action>
-  </phase>
-</check>
-```
+### Output Files
+- [Path to create/modify]
+
+### Requirements
+[Numbered list of what to implement with signatures/details]
 
-## Task Validation
-
-```xml
-<validation>
-  <check>ALL tests for this task EXIST and PASS 100%</check>
-  <check>Implementation matches EXACTLY what task specifies</check>
-  <check>Related acceptance criteria satisfied</check>
-  <check>Full test suite - NO regressions</check>
-</validation>
-
-<check if="ALL pass">
-  <action>Mark task [x] in story file</action>
-  <action>Update File List</action>
-</check>
-
-<check if="ANY fails">
-  <action>DO NOT mark task complete</action>
-  <action>Fix issues first</action>
-</check>
+### Implementation Pattern
+[Code example or reference to existing code]
+
+### Error Handling
+[How to handle errors]
+
+### Done When
+- [ ] [Specific checkable item]
+- [ ] [Another item]
+- [ ] File compiles
+- [ ] Tests pass
 ```
 
-## HALT Conditions
+## Parallel Tasks
+
+When delegating multiple tasks in one message:
+1. Each task gets full context (don't assume @coder remembers)
+2. Clearly separate tasks with headers
+3. No shared state between parallel tasks
+4. Tasks must work on different files
+
+## Task Boundaries
+
+**Good task:**
+- Logically complete unit (service, handler, entity, feature)
+- Clear single responsibility
+- Can be tested independently
+
+**Consider splitting when:**
+- Multiple unrelated responsibilities
+- No logical connection between parts
+
+**Consider combining when:**
+- Tasks too granular to be meaningful
+- Same file, same concern
+
+## Common Patterns
+
+### New Service
+Context: domain entities, repository interface, existing service example
+Requirements: interface, constructor, methods
+Pattern: existing service structure
+
+### New Handler
+Context: service interface, DTOs, existing handler example
+Requirements: handler struct, methods, error mapping
+Pattern: existing handler structure
 
-- Task dependencies not satisfied
-- Additional dependencies need user approval
-- 3 consecutive implementation failures
-- Required configuration missing
+### New Tests
+Context: code to test, existing test examples
+Requirements: test scenarios, mocks
+Pattern: existing test structure
 
-## Definition of Done
+## Methodology
 
-- [ ] All tasks marked [x]
-- [ ] Implementation satisfies every AC
-- [ ] Unit tests added
-- [ ] Integration tests (if required)
-- [ ] All tests pass
-- [ ] File List complete
-- [ ] Change Log updated
+Include in task based on config.yaml:
+- **TDD**: "Write failing test first, then implement"
+- **STUB**: "Create stub first, write tests, then real implementation"

package/src/opencode/skills/jira-integration/SKILL.md

@@ -362,7 +362,7 @@ Agent manages the full development lifecycle:
 | From | To | Trigger | Agent Action |
 |------|----|---------|--------------|
 | backlog | planned | Sprint planning | Update Jira status |
-| planned | in_progress | `/dev-story` starts | Create branch, update status |
+| planned | in_progress | Dev starts work | Create branch, update status |
 | in_progress | review | PR created | Link PR, update status |
 | review | in_progress | Review failed | Update status, add comment |
 | review | qa | Review passed | Update status |

package/src/opencode/skills/prd-validation/SKILL.md

@@ -126,8 +126,8 @@ For each FR in PRD:
 ## Next Steps
 
 After fixing issues:
-- Re-run `/validate prd`
-- If PASS, proceed to `/architecture`
+- Re-run validation
+- If PASS, proceed to architecture design
 ```
 
 ## If QA Artifact Missing

package/src/opencode/skills/requirements-validation/SKILL.md

@@ -127,8 +127,8 @@ Scan for ambiguous language:
 ## Next Steps
 
 After fixing issues:
-- Re-run `/validate requirements`
-- If PASS, proceed to `/prd`
+- Re-run validation
+- If PASS, proceed to PRD creation
 ```
 
 ## Output

package/src/opencode/skills/sprint-planning/SKILL.md

@@ -28,22 +28,22 @@ Check: ls docs/sprint-artifacts/backlog/epic-*.md 2>/dev/null || ls docs/sprint-
 
 | Check | Action if Missing |
 |-------|-------------------|
-| No epics in backlog or sprints | **STOP** → Run `/epics` first to create epics from PRD |
-| Epics exist but no stories | Epics can be planned, stories created later with `/stories` |
-| PRD missing | **STOP** → Run `/prd` first |
+| No epics in backlog or sprints | **STOP** → Create epics from PRD first |
+| Epics exist but no stories | Epics can be planned, stories created later |
+| PRD missing | **STOP** → Create PRD first |
 
 ### Dependency Chain
 
 ```
-/prd → /architecture → /epics → /sprint-plan
-                          ↑
-                    YOU ARE HERE
-                    (need epics first!)
+PRD → Architecture → Epics → Sprint Plan
+                       ↑
+                 YOU ARE HERE
+                 (need epics first!)
 ```
 
 If user asks for sprint planning without epics:
 1. Inform them epics are required
-2. Offer to create epics first with `/epics`
+2. Offer to create epics first
 3. Only proceed with sprint planning after epics exist
 
 ## Sprint Duration

package/src/opencode/skills/translation/SKILL.md

@@ -239,7 +239,7 @@ Kafka, Redis, PostgreSQL, MongoDB
 </workflow>
 ```
 
-### Command: `/export-confluence`
+### Batch Export
 
 Batch export all configured documents: