oh-my-customcode 0.1.2 → 0.1.4

package/templates/agents/manager/updater/refs/rules/SHOULD-hud-statusline.md

@@ -0,0 +1,96 @@
+# [SHOULD] HUD Statusline Rules
+
+> **Priority**: SHOULD - Recommended for visibility
+> **ID**: R012
+
+## Purpose
+
+Display real-time status information during agent operations for improved visibility and progress tracking.
+
+## Format
+
+```
+─── [Agent] {name} | [Progress] {n}/{total} | [Parallel] {count} ───
+```
+
+## Update Triggers
+
+| Trigger | Description |
+|---------|-------------|
+| Agent activation | When an agent starts handling a task |
+| Task start | When a specific task begins execution |
+| Task complete | When a task finishes (success or failure) |
+| Parallel spawn | When parallel instances are created |
+| Parallel complete | When parallel instances finish |
+
+## Components
+
+| Component | Description | Example |
+|-----------|-------------|---------|
+| Agent | Active agent name | `golang-expert` |
+| Model | Model used (for parallel) | `sonnet` |
+| Progress | Task progress (current/total) | `2/5` |
+| Parallel | Parallel instance count | `3` |
+
+## Display Rules
+
+### Single Agent
+
+```
+─── [Agent] creator | [Progress] 1/3 ───
+```
+
+### With Parallel Execution (Model Display REQUIRED)
+
+```
+─── [Agent] secretary | [Progress] 0/4 | [Parallel] 4 ───
+
+Instances:
+  [1] readme-update:sonnet
+  [2] code-review:haiku
+  [3] architecture:opus
+  [4] validation:haiku
+```
+
+### Completion
+
+```
+─── [Agent] supplier | [Progress] 5/5 | [Done] ───
+```
+
+## Implementation
+
+Status updates via stderr to avoid output pollution:
+
+```bash
+echo "─── [Agent] $AGENT | [Progress] $PROGRESS ───" >&2
+```
+
+## Integration with Other Rules
+
+| Rule | Integration |
+|------|-------------|
+| R007 (Agent ID) | HUD complements agent identification |
+| R008 (Tool ID) | HUD shows overall progress, tool ID shows specific operations |
+| R009 (Parallel) | HUD displays parallel instance count |
+
+## When to Display
+
+| Situation | Display HUD |
+|-----------|-------------|
+| Single file operation | No (too brief) |
+| Multi-step task | Yes |
+| Parallel execution | Yes |
+| Long-running operations | Yes |
+
+## Hook Usage
+
+The HUD statusline hook can be triggered via:
+
+```bash
+~/.claude/hooks/hud/update-status.sh <agent> [progress] [parallel_count]
+
+# Examples
+~/.claude/hooks/hud/update-status.sh "creator" "1/3"
+~/.claude/hooks/hud/update-status.sh "secretary" "0/4" "4"
+```

package/templates/agents/manager/updater/refs/rules/SHOULD-interaction.md

@@ -0,0 +1,103 @@
+# [SHOULD] Interaction Rules
+
+> **Priority**: SHOULD - Strongly recommended
+> **Exception**: Emergency or explicit user request
+
+## Response Principles
+
+### 1. Brevity
+```
+✓ Key information first
+✓ Skip unnecessary preamble
+✓ Answer only what's asked
+✗ Over-explanation
+✗ Repetitive confirmation
+```
+
+### 2. Clarity
+```
+✓ Specific expressions
+✓ Unambiguous instructions
+✓ Executable code format
+✗ Abstract descriptions only
+✗ Overuse of "maybe", "probably"
+```
+
+### 3. Transparency
+```
+✓ State actions performed
+✓ Report changes
+✓ Acknowledge uncertainty
+✗ Hide actions
+✗ Present guesses as facts
+```
+
+## Status Report Format
+
+### Start
+```
+[Start] {task name}
+```
+
+### In Progress
+```
+[Progress] {current step} ({n}/{total})
+```
+
+### Complete
+```
+[Done] {task name}
+Result: {summary}
+```
+
+### Failed
+```
+[Failed] {task name}
+Cause: {reason}
+Alternative: {possible solutions}
+```
+
+## Handling Requests
+
+### Clear Request
+→ Execute immediately
+
+### Ambiguous Request
+```
+[Confirm]
+Understood "{request}" as {interpretation}.
+
+Proceed?
+```
+
+### Risky Request
+```
+[Warning]
+This action has {risk factor}.
+
+Continue?
+- Yes: {action to perform}
+- No: Cancel
+```
+
+## Multiple Tasks
+
+### Order
+1. Dependent tasks: Sequential
+2. Independent tasks: Parallel allowed
+
+### Report
+```
+[Task 1/3] Done - {result}
+[Task 2/3] In progress...
+[Task 3/3] Pending
+```
+
+## Long-running Tasks
+
+```
+[In Progress] {task name}
+Elapsed: {time}
+Current: {step}
+Remaining: {work left}
+```

package/templates/agents/manager/updater/refs/rules/SHOULD-memory-integration.md

@@ -0,0 +1,114 @@
+# [SHOULD] Memory Integration Rules
+
+> **Priority**: SHOULD - Recommended for context persistence
+> **ID**: R011
+
+## Purpose
+
+Integrate claude-mem for session context persistence across compaction.
+
+## Requirements
+
+### 1. Save Before Compaction
+```yaml
+trigger: PreCompact hook
+action:
+  - Collect session context (tasks, decisions, key info)
+  - Format with project tag "baekgom-agents"
+  - Store in claude-mem with metadata
+```
+
+### 2. Restore on Session Start
+```yaml
+trigger: SessionStart hook
+action:
+  - Build semantic query with project prefix
+  - Search claude-mem for relevant context
+  - Load and present context to agent
+```
+
+### 3. Project Isolation
+```yaml
+rule: Always include project name in queries
+reason: Prevent cross-contamination between projects
+```
+
+## Storage Format
+
+```yaml
+project: baekgom-agents
+session: {date}-{uuid}
+tags: [session, task, decision]
+content:
+  summary: Brief description of session context
+  tasks_completed: List of completed tasks
+  decisions: Key decisions made
+  open_items: Unfinished work
+```
+
+## Query Pattern
+
+Always include project name in queries:
+- `"baekgom-agents session authentication"`
+- `"baekgom-agents 2025-01-24 bug fix"`
+- `"baekgom-agents agent creation workflow"`
+
+### Effective Queries
+```yaml
+good:
+  - "baekgom-agents implementing oauth" (semantic, project-scoped)
+  - "baekgom-agents 2025-01-24 memory system" (temporal)
+  - "baekgom-agents decision agent architecture" (topic-based)
+
+bad:
+  - "implementing oauth" (missing project scope)
+  - "session context" (too generic)
+```
+
+## Commands
+
+| Command | Description |
+|---------|-------------|
+| `memory:save` | Save current context to claude-mem |
+| `memory:recall` | Search and recall relevant memories |
+
+## Integration with Agents
+
+### memory-keeper Agent
+```
+Responsible for:
+- Executing save/recall operations
+- Managing session metadata
+- Handling PreCompact and SessionStart hooks
+```
+
+### Other Agents
+```
+When to trigger memory:save:
+- Before complex task completion
+- When making significant decisions
+- Before expected context compaction
+```
+
+## Storage Provider
+
+```yaml
+provider: claude-mem
+collection: claude_memories
+project_tag: baekgom-agents
+archive_path: ~/.claude-mem/archives/
+```
+
+## Error Handling
+
+```yaml
+on_save_failure:
+  - Log error
+  - Continue without blocking main task
+  - Notify user of memory save failure
+
+on_recall_failure:
+  - Log error
+  - Continue with available context
+  - Notify user that memory recall failed
+```

package/templates/agents/manager/updater/refs/rules/SHOULD-pipeline-mode.md

@@ -0,0 +1,165 @@
+# [SHOULD] Pipeline Mode Rules
+
+> **Priority**: SHOULD - Recommended for sequential workflows
+> **ID**: R014
+
+## Purpose
+
+Enable declarative, sequential multi-step workflows with variable passing between steps.
+
+## Pipeline Definition Format
+
+```yaml
+name: pipeline-name
+description: What this pipeline does
+
+inputs:
+  - name: input_name
+    required: true
+    description: Input description
+
+steps:
+  - id: step_1
+    agent: agent-name
+    action: action_name
+    input: { key: "${input_name}" }
+    output: result_variable
+
+  - id: step_2
+    agent: another-agent
+    action: action_name
+    input: { data: "${result_variable}" }
+    output: final_result
+```
+
+## Variable Syntax
+
+- `${variable}`: Reference input or previous step output
+- `${step_id.field}`: Reference specific field from step output
+- `${inputs.name}`: Explicit input reference
+
+## Execution Rules
+
+### Sequential by Default
+
+Steps execute in order. Each step waits for previous step completion.
+
+```
+Step 1 → complete → Step 2 → complete → Step 3
+```
+
+### Output Passing
+
+Output from step N is available to step N+1 and beyond.
+
+```yaml
+steps:
+  - id: analyze
+    output: analysis    # Creates ${analysis}
+
+  - id: review
+    input: { data: "${analysis}" }  # Uses ${analysis}
+```
+
+### Error Handling
+
+```yaml
+error_handling:
+  default: stop              # stop | continue | retry
+
+steps:
+  - id: risky_step
+    on_error: continue       # Override for this step
+    retry:
+      max_attempts: 3
+      delay: 1000
+```
+
+## Pipeline Commands
+
+| Command | Description |
+|---------|-------------|
+| `pipeline:run <name>` | Execute a pipeline |
+| `pipeline:list` | List available pipelines |
+| `pipeline:status` | Show running pipeline status |
+
+## Pipeline Locations
+
+```
+pipelines/
+├── index.yaml              # Pipeline registry
+├── templates/              # Pipeline templates
+│   └── pipeline-template.yaml
+└── examples/               # Example pipelines
+    └── code-review.yaml
+```
+
+## Example Pipeline
+
+### code-review.yaml
+
+```yaml
+name: code-review
+description: Full code review workflow
+
+inputs:
+  - name: file_path
+    required: true
+
+steps:
+  - id: detect_language
+    agent: secretary
+    action: detect_language
+    input: { file: "${file_path}" }
+    output: language
+
+  - id: analyze
+    agent: "${language}-expert"
+    action: analyze
+    input: { file: "${file_path}" }
+    output: analysis
+
+  - id: security_check
+    agent: qa-lead
+    action: security_review
+    input: { findings: "${analysis}" }
+    output: security
+
+  - id: report
+    agent: documenter
+    action: summarize
+    input:
+      analysis: "${analysis}"
+      security: "${security}"
+    output: report
+```
+
+## Execution Output
+
+```
+[Pipeline] code-review
+├── [1/4] detect_language: ✓ Go detected
+├── [2/4] analyze: ✓ 5 issues found
+├── [3/4] security_check: ✓ No vulnerabilities
+└── [4/4] report: ✓ Summary generated
+
+[Complete] Pipeline finished successfully
+Output: ${report}
+```
+
+## Secretary Integration
+
+Secretary can:
+1. Load and validate pipeline definitions
+2. Execute pipelines step-by-step
+3. Pass variables between steps
+4. Report progress and results
+5. Handle errors according to policy
+
+## Benefits
+
+1. **Reproducibility**: Same workflow, consistent results
+2. **Composability**: Build complex workflows from simple steps
+3. **Visibility**: Clear step-by-step progress
+4. **Error Recovery**: Defined error handling per step
+5. **Reusability**: Save and reuse common workflows

package/templates/agents/manager/updater/refs/rules/index.yaml

@@ -0,0 +1,125 @@
+# Baekgom Agents - Global Rules
+# Priority: MUST > SHOULD > MAY
+
+rules:
+  # MUST - Never violate
+  - id: R000
+    name: language-policy
+    title: Language & Delegation Policy
+    path: ./MUST-language-policy.md
+    priority: MUST
+    scope: all
+
+  - id: R001
+    name: safety
+    title: Safety Rules
+    path: ./MUST-safety.md
+    priority: MUST
+    scope: all
+
+  - id: R002
+    name: permissions
+    title: Permission Rules
+    path: ./MUST-permissions.md
+    priority: MUST
+    scope: all
+
+  # SHOULD - Strongly recommended
+  - id: R003
+    name: interaction
+    title: Interaction Rules
+    path: ./SHOULD-interaction.md
+    priority: SHOULD
+    scope: all
+
+  - id: R004
+    name: error-handling
+    title: Error Handling Rules
+    path: ./SHOULD-error-handling.md
+    priority: SHOULD
+    scope: all
+
+  - id: R007
+    name: agent-identification
+    title: Agent Identification Rules
+    path: ./SHOULD-agent-identification.md
+    priority: SHOULD
+    scope: all
+
+  - id: R008
+    name: tool-identification
+    title: Tool Usage Identification Rules
+    path: ./SHOULD-tool-identification.md
+    priority: SHOULD
+    scope: all
+
+  # MAY - Optional
+  - id: R005
+    name: optimization
+    title: Optimization Guide
+    path: ./MAY-optimization.md
+    priority: MAY
+    scope: all
+
+  - id: R009
+    name: parallel-execution
+    title: Parallel Execution Rules
+    path: ./MAY-parallel-execution.md
+    priority: MAY
+    scope: agents
+
+  # Agent Design - MUST
+  - id: R006
+    name: agent-design
+    title: Agent Design Rules
+    path: ./MUST-agent-design.md
+    priority: MUST
+    scope: agents
+
+  # Orchestrator Coordination - MUST
+  - id: R010
+    name: orchestrator-coordination
+    title: Orchestrator Coordination Rules
+    path: ./MUST-orchestrator-coordination.md
+    priority: MUST
+    scope: agents
+
+  # Memory Integration - SHOULD
+  - id: R011
+    name: memory-integration
+    title: Memory Integration Rules
+    path: ./SHOULD-memory-integration.md
+    priority: SHOULD
+    scope: all
+
+  # HUD Statusline - SHOULD
+  - id: R012
+    name: hud-statusline
+    title: HUD Statusline Rules
+    path: ./SHOULD-hud-statusline.md
+    priority: SHOULD
+    scope: all
+
+  # Ecomode - SHOULD
+  - id: R013
+    name: ecomode
+    title: Ecomode Rules
+    path: ./SHOULD-ecomode.md
+    priority: SHOULD
+    scope: all
+
+  # Pipeline Mode - SHOULD
+  - id: R014
+    name: pipeline-mode
+    title: Pipeline Mode Rules
+    path: ./SHOULD-pipeline-mode.md
+    priority: SHOULD
+    scope: orchestrator
+
+  # Intent Transparency - MUST
+  - id: R015
+    name: intent-transparency
+    title: Intent Transparency Rules
+    path: ./MUST-intent-transparency.md
+    priority: MUST
+    scope: orchestrator

package/templates/agents/orchestrator/dev-lead/AGENT.md

@@ -94,6 +94,53 @@ Dev Lead:
 Result: 3 reviews in parallel
 ```
 
+## Sub-agent Model Specification
+
+Use Task tool's `model` parameter for cost/performance optimization:
+
+### Model Selection
+
+| Model | Use Case | Cost |
+|-------|----------|------|
+| `opus` | Architecture analysis, complex refactoring | High |
+| `sonnet` | Code implementation, standard review (default) | Medium |
+| `haiku` | File search, quick validation | Low |
+
+### Task Call Examples
+
+```
+# Complex architecture analysis
+Task(
+  subagent_type: "general-purpose",
+  prompt: "Analyze module dependencies and suggest improvements",
+  model: "opus"
+)
+
+# Standard code review
+Task(
+  subagent_type: "general-purpose",
+  prompt: "Review Go code in src/handlers/",
+  model: "sonnet"
+)
+
+# Quick file search
+Task(
+  subagent_type: "Explore",
+  prompt: "Find all files importing auth package",
+  model: "haiku"
+)
+```
+
+### Expert Agent Model Mapping
+
+| Agent Type | Recommended Model | Reason |
+|------------|-------------------|--------|
+| Language experts | `sonnet` | Balanced code generation |
+| Frontend experts | `sonnet` | Component implementation |
+| Backend experts | `sonnet` | API implementation |
+| Architecture review | `opus` | Deep reasoning needed |
+| Quick validation | `haiku` | Fast response |
+
 ## Output Format
 
 ```

package/templates/agents/orchestrator/secretary/AGENT.md

@@ -99,6 +99,55 @@ Secretary:
 Result: 3 agents created in parallel
 ```
 
+## Sub-agent Model Specification
+
+Use Task tool's `model` parameter to optimize cost and performance:
+
+### Model Selection
+
+| Model | Use Case | Cost |
+|-------|----------|------|
+| `opus` | Complex analysis, architecture | High |
+| `sonnet` | General development (default) | Medium |
+| `haiku` | Simple tasks, file search | Low |
+| `inherit` | Match parent model | Varies |
+
+### Task Call Examples
+
+```
+# Complex task requiring deep reasoning
+Task(
+  subagent_type: "general-purpose",
+  prompt: "Analyze agent dependencies and suggest improvements",
+  model: "opus"
+)
+
+# Standard manager task
+Task(
+  subagent_type: "general-purpose",
+  prompt: "Create new golang-expert agent following creator workflow",
+  model: "sonnet"
+)
+
+# Simple file operation
+Task(
+  subagent_type: "general-purpose",
+  prompt: "Search for all AGENT.md files",
+  model: "haiku"
+)
+```
+
+### Manager Agent Model Mapping
+
+| Agent | Recommended Model | Reason |
+|-------|-------------------|--------|
+| creator | `sonnet` | File generation, balanced |
+| updater | `sonnet` | External sync, web fetch |
+| supplier | `haiku` | File scan, validation |
+| gitnerd | `sonnet` | Commit message quality |
+| sync-checker | `haiku` | Fast verification |
+| sauron | `sonnet` | Multi-round verification |
+
 ## Error Handling
 
 ```yaml