oh-my-claude-sisyphus 3.5.1 → 3.5.5

package/skills/learn-about-omc/SKILL.md

@@ -0,0 +1,128 @@
+---
+name: learn-about-omc
+description: Analyze your OMC usage patterns and get personalized recommendations
+---
+
+# Learn About OMC
+
+Analyzes your oh-my-claudecode usage and provides tailored recommendations to improve your workflow.
+
+## What It Does
+
+1. Reads token tracking from `~/.omc/state/token-tracking.jsonl`
+2. Reads session history from `.omc/state/session-history.json`
+3. Analyzes agent usage patterns
+4. Identifies underutilized features
+5. Recommends configuration changes
+
+## Implementation
+
+### Step 1: Gather Data
+
+```bash
+# Check for token tracking data
+TOKEN_FILE="$HOME/.omc/state/token-tracking.jsonl"
+SESSION_FILE=".omc/state/session-history.json"
+CONFIG_FILE="$HOME/.claude/.omc-config.json"
+
+echo "📊 Analyzing OMC Usage..."
+echo ""
+
+# Check what data is available
+HAS_TOKENS=false
+HAS_SESSIONS=false
+HAS_CONFIG=false
+
+if [[ -f "$TOKEN_FILE" ]]; then
+  HAS_TOKENS=true
+  TOKEN_COUNT=$(wc -l < "$TOKEN_FILE")
+  echo "Token records found: $TOKEN_COUNT"
+fi
+
+if [[ -f "$SESSION_FILE" ]]; then
+  HAS_SESSIONS=true
+  SESSION_COUNT=$(cat "$SESSION_FILE" | jq '.sessions | length' 2>/dev/null || echo "0")
+  echo "Sessions found: $SESSION_COUNT"
+fi
+
+if [[ -f "$CONFIG_FILE" ]]; then
+  HAS_CONFIG=true
+  DEFAULT_MODE=$(cat "$CONFIG_FILE" | jq -r '.defaultExecutionMode // "not set"')
+  echo "Default execution mode: $DEFAULT_MODE"
+fi
+```
+
+### Step 2: Analyze Agent Usage (if token data exists)
+
+```bash
+if [[ "$HAS_TOKENS" == "true" ]]; then
+  echo ""
+  echo "TOP AGENTS BY USAGE:"
+  cat "$TOKEN_FILE" | jq -r '.agentName // "main"' | sort | uniq -c | sort -rn | head -10
+
+  echo ""
+  echo "MODEL DISTRIBUTION:"
+  cat "$TOKEN_FILE" | jq -r '.modelName' | sort | uniq -c | sort -rn
+fi
+```
+
+### Step 3: Generate Recommendations
+
+Based on patterns found, output recommendations:
+
+**If high Opus usage (>40%) and no ecomode:**
+- "Consider using ecomode for routine tasks to save tokens"
+
+**If no pipeline usage:**
+- "Try /pipeline for code review workflows"
+
+**If no security-reviewer usage:**
+- "Use security-reviewer after auth/API changes"
+
+**If defaultExecutionMode not set:**
+- "Set defaultExecutionMode in /omc-setup for consistent behavior"
+
+### Step 4: Output Report
+
+Format a nice summary with:
+- Token summary (total, by model)
+- Top agents used
+- Underutilized features
+- Personalized recommendations
+
+### Graceful Degradation
+
+If no data found:
+```
+📊 Limited Usage Data Available
+
+No token tracking found. To enable tracking:
+1. Ensure ~/.omc/state/ directory exists
+2. Run any OMC command to start tracking
+
+Tip: Run /omc-setup to configure OMC properly.
+```
+
+## Example Output
+
+```
+📊 Your OMC Usage Analysis
+
+TOKEN SUMMARY:
+- Total records: 1,234
+- By Model: opus 45%, sonnet 40%, haiku 15%
+
+TOP AGENTS:
+1. executor (234 uses)
+2. architect (89 uses)
+3. explore (67 uses)
+
+UNDERUTILIZED FEATURES:
+- ecomode: 0 uses (could save ~30% on routine tasks)
+- pipeline: 0 uses (great for review workflows)
+
+RECOMMENDATIONS:
+1. Set defaultExecutionMode: "ecomode" to save tokens
+2. Try /pipeline review for PR reviews
+3. Use explore agent before architect to save context
+```

package/skills/omc-setup/SKILL.md

@@ -266,30 +266,13 @@ echo "Default execution mode set to: USER_CHOICE"
 
 **Note**: This preference ONLY affects generic keywords ("fast", "parallel"). Explicit keywords ("ulw", "eco") always override this preference.
 
-## Step 4: Install Analytics CLI Dependencies
-
-The analytics CLI requires additional dependencies for beautiful output formatting.
-
-```bash
-# Check if we're in the oh-my-claudecode directory
-if [ -f "package.json" ] && grep -q "oh-my-claude-sisyphus" package.json 2>/dev/null; then
-  echo "Installing analytics CLI dependencies..."
-  npm install --silent 2>&1 | grep -v "npm WARN" || echo "Dependencies installed"
-  echo "Analytics CLI ready: omc"
-else
-  echo "Skipping analytics dependencies (not in oh-my-claudecode directory)"
-fi
-```
-
-**Note:** This step only runs when setup is executed from within the oh-my-claudecode project directory.
-
-## Step 5: Verify Plugin Installation
+## Step 4: Verify Plugin Installation
 
 ```bash
 grep -q "oh-my-claudecode" ~/.claude/settings.json && echo "Plugin verified" || echo "Plugin NOT found - run: claude /install-plugin oh-my-claudecode"
 ```
 
-## Step 6: Offer MCP Server Configuration
+## Step 5: Offer MCP Server Configuration
 
 MCP servers extend Claude Code with additional tools (web search, GitHub, etc.).
 
@@ -302,7 +285,7 @@ If yes, invoke the mcp-setup skill:
 
 If no, skip to next step.
 
-## Step 7: Detect Upgrade from 2.x
+## Step 6: Detect Upgrade from 2.x
 
 Check if user has existing configuration:
 ```bash
@@ -312,7 +295,7 @@ ls ~/.claude/commands/ralph-loop.md 2>/dev/null || ls ~/.claude/commands/ultrawo
 
 If found, this is an upgrade from 2.x.
 
-## Step 8: Show Welcome Message
+## Step 7: Show Welcome Message
 
 ### For New Users:
 
@@ -343,10 +326,6 @@ Combine them: "ralph ulw: migrate the database"
 MCP SERVERS:
 Run /oh-my-claudecode:mcp-setup to add tools like web search, GitHub, etc.
 
-ANALYTICS CLI:
-Track token usage and costs with: omc | omc stats | omc cost | omc sessions | omc agents
-Run 'omc' alone to see everything (session stats, cost, agents, cache efficiency)
-
 HUD STATUSLINE:
 The status bar now shows OMC state. Restart Claude Code to see it.
 
@@ -359,14 +338,14 @@ That's it! Just use Claude Code normally.
 OMC Setup Complete! (Upgraded from 2.x)
 
 GOOD NEWS: Your existing commands still work!
-- /ralph, /ultrawork, /planner, etc. all still function
+- /ralph, /ultrawork, /plan, etc. all still function
 
 WHAT'S NEW in 3.0:
 You no longer NEED those commands. Everything is automatic now:
 - Just say "don't stop until done" instead of /ralph
 - Just say "fast" or "parallel" instead of /ultrawork
-- Just say "plan this" instead of /planner
-- Just say "stop" instead of /cancel-ralph
+- Just say "plan this" instead of /plan
+- Just say "stop" instead of /cancel
 
 MAGIC KEYWORDS (power-user shortcuts):
 | Keyword | Same as old... | Example |
@@ -375,11 +354,7 @@ MAGIC KEYWORDS (power-user shortcuts):
 | ralplan | /ralplan | "ralplan this feature" |
 | ulw | /ultrawork | "ulw refactor API" |
 | eco | (new!) | "eco fix all errors" |
-| plan | /planner | "plan the endpoints" |
-
-ANALYTICS CLI:
-Track token usage and costs with: omc | omc stats | omc cost | omc sessions | omc agents
-Run 'omc' alone to see everything (session stats, cost, agents, cache efficiency)
+| plan | /plan | "plan the endpoints" |
 
 HUD STATUSLINE:
 The status bar now shows OMC state. Restart Claude Code to see it.
@@ -387,7 +362,7 @@ The status bar now shows OMC state. Restart Claude Code to see it.
 Your workflow won't break - it just got easier!
 ```
 
-## Step 9: Ask About Starring Repository
+## Step 8: Ask About Starring Repository
 
 First, check if `gh` CLI is available and authenticated:
 

package/skills/ralplan/SKILL.md

@@ -3,269 +3,38 @@ name: ralplan
 description: Iterative planning with Planner, Architect, and Critic until consensus
 ---
 
-# Ralplan Skill
+# Ralplan Command
 
-## Overview
+**This is an alias for `/plan --consensus`**
 
-Ralplan orchestrates three specialized agents—Planner, Architect, and Critic—in an iterative loop until consensus is reached on a comprehensive work plan. This skill ensures plans are strategically sound, architecturally valid, and thoroughly reviewed before execution.
+Ralplan orchestrates three specialized agents—Planner, Architect, and Critic—in an iterative loop until consensus is reached on a comprehensive work plan.
 
-## The Planning Triad
+## Usage
 
-Three agents collaborate in structured phases to validate and refine work plans:
-
-| Agent | Role | Output |
-|-------|------|--------|
-| **Planner** | Strategic Planner | Creates/refines the work plan |
-| **Architect** | Strategic Advisor | Answers questions, validates architecture |
-| **Critic** | Ruthless Reviewer | Critiques and identifies gaps |
-
-## The Iteration Loop
-
-```
-┌─────────────────────────────────────────────────────────────────┐
-│                       RALPLAN LOOP                              │
-│                                                                 │
-│    ┌──────────────┐                                             │
-│    │   PLANNER    │◄────────────────────────────────┐           │
-│    │   (Plans)    │                                 │           │
-│    └──────┬───────┘                                 │           │
-│           │                                         │           │
-│           ▼                                         │           │
-│    ┌──────────────┐     Questions?    ┌───────────┐ │           │
-│    │   Has open   │─────────────────► │ ARCHITECT │ │           │
-│    │  questions?  │                   │ (Advises) │ │           │
-│    └──────┬───────┘                   └─────┬─────┘ │           │
-│           │                                 │       │           │
-│           │ No questions                    │       │           │
-│           ▼                                 ▼       │           │
-│    ┌──────────────┐                  ┌──────────┐   │           │
-│    │    CRITIC    │◄─────────────────│ Answers  │   │           │
-│    │  (Reviews)   │                  └──────────┘   │           │
-│    └──────┬───────┘                                 │           │
-│           │                                         │           │
-│           ▼                                         │           │
-│    ┌──────────────┐     REJECT      ┌──────────────┐│           │
-│    │   Verdict?   │─────────────────►│  Feedback   ││           │
-│    └──────┬───────┘                  │ to Planner  │┘           │
-│           │                          └─────────────┘            │
-│           │ OKAY                                                │
-│           ▼                                                     │
-│    ┌──────────────────────────────────────────────────────────┐ │
-│    │                  PLAN APPROVED                           │ │
-│    │           Ready for /ralph execution                     │ │
-│    └──────────────────────────────────────────────────────────┘ │
-└─────────────────────────────────────────────────────────────────┘
 ```
-
-## State Management
-
-Ralplan maintains persistent state in `.omc/ralplan-state.json` to track progress and enable recovery across interruptions:
-
-```json
-{
-  "active": true,
-  "mode": "ralplan",
-  "iteration": 1,
-  "max_iterations": 5,
-  "plan_path": ".omc/plans/[feature].md",
-  "current_phase": "planner_planning",
-  "started_at": "ISO-timestamp",
-  "task_description": "[original task]"
-}
+/oh-my-claudecode:ralplan [task]
 ```
 
-**Phases**: `planner_planning` → `architect_consultation` → `critic_review` → `handling_verdict` → `complete`
-
-## Plan Mode Interaction (CRITICAL)
-
-**When ralplan is invoked while Plan mode is active:**
-
-The Planner agent has a PHASE 3.5 (CONFIRMATION) that normally requires user confirmation before proceeding. **This MUST be bypassed within ralplan because the Critic serves as the reviewer, not the user.**
-
-### Mandatory Constraints
-
-| Constraint | Rationale |
-|------------|-----------|
-| **Planner MUST skip PHASE 3.5 confirmation** | Critic reviews the plan, not the user |
-| **Critic MUST run before any plan approval** | No plan is approved without Critic verdict |
-| **DO NOT exit plan mode until Critic has rendered verdict** | Premature exit skips the review step |
-| **Log `[RALPLAN] Critic review required before approval`** | Observability for debugging |
-
-### Flow When Plan Mode Is Active
-
-```
-1. Orchestrator invokes ralplan skill
-2. Log: [RALPLAN] Starting - Critic review will be required
-3. Spawn Planner in DIRECT PLANNING MODE (no interview, no confirmation)
-4. Planner outputs PLAN_READY: <path>
-5. Log: [RALPLAN] Plan ready, invoking Critic for review
-6. Spawn Critic with plan path  <-- MUST HAPPEN
-7. Critic renders OKAY or REJECT
-8. ONLY THEN can plan be approved or refined
-```
-
-**CRITICAL: The Critic invocation is MANDATORY. If the Planner completes and signals plan ready, the orchestrator MUST invoke the Critic before any form of plan approval or user confirmation.**
-
----
-
-## Execution Protocol
-
-### Initialization
-
-The skill begins by establishing the planning environment:
-
-1. Create `.omc/plans/` directory if it doesn't exist
-2. Read task description from user input
-3. Create `ralplan-state.json` with initial values:
-   - `active: true`
-   - `iteration: 0`
-   - `max_iterations: 5`
-   - `current_phase: "planner_planning"`
-   - `started_at`: Current ISO timestamp
-   - `task_description`: User's task description
-
-### Planner Planning Phase
-
-The Planner creates an initial plan based on task context:
-
-- Invoke Planner in **direct planning mode** (bypassing interview since task context is pre-gathered)
-- **CRITICAL: Planner MUST skip PHASE 3.5 (CONFIRMATION)** - the Critic will review, not the user
-- Planner receives task context directly without preliminary questioning
-- Planner mandatorily consults with Metis for gap detection
-- Planner generates plan directly to `.omc/plans/[feature-name].md`
-- Plan includes: requirements summary, concrete acceptance criteria, specific implementation steps with file references, risk identification with mitigations, and verification steps
-- Signal completion with `PLAN_READY: .omc/plans/[filename].md`
-- **After PLAN_READY: DO NOT approve or confirm the plan. Proceed to Critic Review.**
-- Extract plan path from completion signal and update state
-- Log: `[RALPLAN] Critic review required before approval`
-
-### Architect Consultation (Conditional)
-
-The Architect provides strategic guidance in two scenarios:
-
-1. **After Planner**: If Planner raises architectural questions needing strategic input
-2. **After Critic rejection**: If Critic identifies questions requiring expert guidance
+## What It Does
 
-When invoked, the Architect receives file paths to read for analysis, not summaries. This enables thorough examination of the existing codebase context before providing recommendations.
+Invokes the plan skill with --consensus mode, which:
+1. Creates initial plan with Planner agent
+2. Consults Architect for architectural questions
+3. Reviews with Critic agent
+4. Iterates until Critic approves (max 5 iterations)
 
-### Critic Review (MANDATORY - CANNOT BE SKIPPED)
+## Implementation
 
-**CRITICAL: This phase MUST execute. The Critic is the gatekeeper for plan approval.**
+When this skill is invoked, immediately invoke the plan skill with consensus mode:
 
-Log at start: `[RALPLAN] Invoking Critic for plan review`
-
-The Critic examines the plan against quality standards:
-
-- Critic receives the plan file path (per its design)
-- Critic conducts thorough review of plan completeness and feasibility
-- Critic emits verdict: either `OKAY` (approval) or `REJECT` with specific issues
-
-**Enforcement Rules:**
-1. If Planner signals PLAN_READY, Critic MUST be invoked immediately
-2. DO NOT exit ralplan loop before Critic verdict
-3. DO NOT request user confirmation before Critic verdict
-4. DO NOT trigger any plan mode exit behavior before Critic verdict
-
-Log after Critic completes: `[RALPLAN] Critic verdict: <OKAY|REJECT>`
-
-### Verdict Handling and Iteration
-
-Based on Critic's verdict, the skill either approves the plan or continues iteration:
-
-**If verdict is OKAY:**
-- Mark plan as approved
-- Log approval with iteration count
-- Prepare plan for execution with `/oh-my-claudecode:ralph` or manual orchestration
-- Set state `active: false, current_phase: "complete"`
-
-**If verdict is REJECT:**
-- Extract Critic feedback with specific issues
-- Increment iteration counter
-- If `iteration >= max_iterations` (5):
-  - Force approval with warning about unresolved concerns
-  - Recommend manual review before execution
-- Otherwise:
-  - Feed Critic feedback back to Planner
-  - Return to Planner Planning phase for refinement
-
-## Iteration Rules
-
-| Rule | Description |
-|------|-------------|
-| **Max 5 iterations** | Safety limit prevents infinite loops |
-| **Planner owns plan** | Only Planner writes to plan file |
-| **Architect provides wisdom** | Architect reads and advises, never modifies |
-| **Critic has final say** | Plan approved only when Critic signals OKAY |
-| **Feedback is specific** | Each rejection includes actionable improvements |
-| **State persists** | Progress survives session interruptions |
-
-## Quality Gates
-
-The orchestrator must verify these gates before invoking Critic for each review:
-
-1. **Plan file exists** at the path specified in state
-2. **File references are valid** - Verify all mentioned files exist in codebase
-3. **Acceptance criteria are concrete** - No vague "improve" or "optimize" without measurable metrics
-4. **No ambiguous language** - Each task clearly specifies what to do
-
-If any gate fails, return to Planner with specific failure feedback for remediation.
-
-## Agent Communication Protocol
-
-### Planner to Architect Questions
-
-```
-ARCHITECT_QUESTION:
-- Topic: [Architecture/Performance/Security/Pattern]
-- Context: [What we're planning]
-- Files to examine: [specific paths]
-- Specific Question: [What we need answered]
-```
-
-### Architect to Planner Answers
-
-```
-ARCHITECT_ANSWER:
-- Topic: [Matching topic]
-- Analysis: [What Architect found after reading files]
-- Recommendation: [Specific guidance]
-- Trade-offs: [What to consider]
-- References: [file:line citations from codebase]
 ```
-
-### Critic to Planner Feedback
-
+Invoke Skill: plan --consensus {{ARGUMENTS}}
 ```
-CRITIC_FEEDBACK:
-- Verdict: REJECT
-- Critical Issues:
-  1. [Issue with specific fix required]
-  2. [Issue with specific fix required]
-- Minor Issues:
-  1. [Nice to fix]
-- Questions for Architect (if any):
-  1. [Architectural question needing expert input]
-```
-
-## Cancellation
-
-To stop an active ralplan session:
-
-- Use `/oh-my-claudecode:cancel-ralph` (automatically detects ralplan via state file)
-- Or manually delete `.omc/ralplan-state.json`
-
-## Skill Workflow
 
-1. **Initialize state** and log: `[RALPLAN Iteration 0/5] Initializing...`
-2. **Parse task** from user input
-3. **Spawn Planner** in direct planning mode (SKIP Planner's PHASE 3.5 confirmation)
-4. **Wait for PLAN_READY** signal from Planner
-5. **Log:** `[RALPLAN] Critic review required before approval`
-6. **Invoke Critic** with plan file path (MANDATORY - CANNOT BE SKIPPED)
-7. **Log:** `[RALPLAN] Critic verdict: <verdict>`
-8. **Handle verdict** - if REJECT, loop back to step 3 with feedback
-9. **Complete** ONLY when Critic approves or max iterations reached with warnings
+Pass all arguments to the plan skill. The plan skill handles all consensus logic, state management, and iteration.
 
-**HARD RULE:** Steps 5-7 are NON-NEGOTIABLE. No plan approval, user confirmation, or plan mode exit can occur before the Critic has rendered its verdict. This prevents the plan mode confirmation flow from short-circuiting the ralplan review loop.
+## See Also
 
-The iterative loop refines the plan until it meets the rigorous standards of all three agents, ensuring comprehensive, architecturally sound work plans ready for execution.
+- `/plan` - Base planning skill with all modes
+- `/plan --consensus` - Direct invocation of consensus mode
+- `/cancel` - Cancel active planning session

package/skills/research/SKILL.md

@@ -145,7 +145,7 @@ Pending stages: {{PENDING_STAGES}}
 1. **Max Iterations:** 10 (configurable)
 2. **Continue until:** Promise tag emitted OR max iterations
 3. **State tracking:** Persist after each stage completion
-4. **Cancellation:** `/cancel-research` or "stop", "cancel"
+4. **Cancellation:** `/oh-my-claudecode:cancel` or "stop", "cancel"
 
 ### AUTO Mode Example
 
@@ -480,7 +480,7 @@ Optional settings in `.claude/settings.json`:
 ## Cancellation
 
 ```
-/oh-my-claudecode:cancel-research
+/oh-my-claudecode:cancel
 ```
 
 Or say: "stop research", "cancel research", "abort"

package/skills/ultrapilot/SKILL.md

@@ -318,7 +318,7 @@ Optional settings in `.claude/settings.json`:
 ## Cancellation
 
 ```
-/oh-my-claudecode:cancel-ultrapilot
+/oh-my-claudecode:cancel
 ```
 
 Or say: "stop", "cancel ultrapilot", "abort"

package/skills/ultraqa/SKILL.md

@@ -107,7 +107,7 @@ Track state in `.omc/ultraqa-state.json`:
 
 ## Cancellation
 
-User can cancel with `/oh-my-claudecode:cancel-ultraqa` which clears the state file.
+User can cancel with `/oh-my-claudecode:cancel` which clears the state file.
 
 ## Important Rules
 

package/commands/cancel-autopilot.md

@@ -1,35 +0,0 @@
----
-description: Cancel active autopilot session
----
-
-# Cancel Autopilot
-
-[CANCELLING AUTOPILOT]
-
-You are cancelling the active autopilot session.
-
-## Action
-
-1. Call the cancel function to clean up state
-2. Report what was cancelled
-3. Show preserved progress
-
-## Steps
-
-1. Check if autopilot is active
-2. Clean up Ralph/UltraQA if active
-3. Preserve autopilot state for resume
-4. Report status
-
-## Arguments
-
-{{ARGUMENTS}}
-
-If `--clear` is passed, completely clear all state instead of preserving.
-
-## Output
-
-Report:
-- What phase was cancelled
-- What modes were cleaned up
-- How to resume

package/commands/cancel-ecomode.md

@@ -1,71 +0,0 @@
----
-description: Cancel active Ecomode mode (deprecated - use /oh-my-claudecode:cancel instead)
----
-
-# Cancel Ecomode
-
-[ECOMODE CANCELLED]
-
-**DEPRECATION NOTICE:** This command is deprecated. Use `/oh-my-claudecode:cancel` instead, which intelligently detects and cancels any active mode including ecomode.
-
-The unified cancel command is safer because it:
-- Checks if ecomode is linked to Ralph and handles both
-- Prevents orphaned state files
-- Provides consistent cancellation experience
-
-## Legacy Behavior
-
-If you still use this command, it will:
-
-1. Check if ecomode is linked to Ralph
-2. If linked → Warn and suggest using `/oh-my-claudecode:cancel-ralph`
-3. If standalone → Cancel ecomode only
-
-## Arguments
-
-{{ARGUMENTS}}
-
-## Recommended Action
-
-Use the unified cancel command:
-```
-/oh-my-claudecode:cancel
-```
-
-This will detect ecomode and cancel it properly, along with any linked modes.
-
-## Implementation
-
-If you must cancel ecomode directly:
-
-```bash
-# Check if linked to ralph
-LINKED=$(cat .omc/ecomode-state.json 2>/dev/null | jq -r '.linked_to_ralph // false')
-
-if [[ "$LINKED" == "true" ]]; then
-  echo "Warning: Ecomode is linked to Ralph."
-  echo "Use /oh-my-claudecode:cancel to cancel both modes."
-  exit 1
-fi
-
-# Cancel standalone ecomode
-mkdir -p .omc ~/.claude
-rm -f .omc/ecomode-state.json
-rm -f ~/.claude/ecomode-state.json
-
-echo "Ecomode cancelled. Token-efficient execution mode deactivated."
-```
-
-## Migration
-
-Replace:
-```bash
-/oh-my-claudecode:cancel-ecomode
-```
-
-With:
-```bash
-/oh-my-claudecode:cancel
-```
-
-The new unified cancel is smarter and safer.