moflo 4.8.9 → 4.8.11

package/.claude/skills/fl/SKILL.md

@@ -13,7 +13,7 @@ Research, create tickets for, and execute GitHub issues automatically.
 ## Usage
 
 ```
-/flo <issue-number>                   # Full workflow with SWARM (default)
+/flo <issue-number>                   # Full workflow in NORMAL mode (default)
 /flo -t <issue-number>                # Ticket only: research and update ticket, then STOP
 /flo -t <title>                       # Create a NEW ticket with description, acceptance criteria, test cases
 /flo --ticket <issue-number|title>    # Same as -t
@@ -26,8 +26,8 @@ Also available as `/fl` (shorthand alias).
 ### Execution Mode (how work is done)
 
 ```
-/flo 123                              # SWARM mode (default) - multi-agent coordination
-/flo -s 123                           # SWARM mode (explicit)
+/flo 123                              # NORMAL mode (default) - single-agent execution
+/flo -s 123                           # SWARM mode - multi-agent coordination
 /flo --swarm 123                      # Same as -s
 /flo -h 123                           # HIVE-MIND mode - consensus-based coordination
 /flo --hive 123                       # Same as -h
@@ -58,22 +58,27 @@ Also available as `/fl` (shorthand alias).
 ### Combined Examples
 
 ```
-/flo 123                              # Swarm + full workflow (default) - includes ALL tests
+/flo 123                              # Normal + full workflow (default) - includes ALL tests
 /flo 42                               # If #42 is epic, processes stories sequentially
-/flo -t 123                           # Swarm + ticket only (no implementation)
+/flo -s 123                           # Swarm + full workflow (multi-agent coordination)
+/flo -t 123                           # Normal + ticket only (no implementation)
 /flo -h -t 123                        # Hive-mind + ticket only
-/flo -n -r 123                        # Normal + research only
+/flo -s -r 123                        # Swarm + research only
 /flo --swarm --ticket 123             # Explicit swarm + ticket only
-/flo -n 123                           # Normal + full workflow (still runs all tests)
+/flo -n 123                           # Normal (explicit, same as default)
 ```
 
-## SWARM IS MANDATORY BY DEFAULT
+## NORMAL MODE IS THE DEFAULT
 
-Even if a task "looks simple", you MUST use swarm coordination unless
-the user explicitly passes -n/--normal. "Simple" is a trap. Tasks have
-hidden complexity. Swarm catches it.
+By default, /flo runs in NORMAL mode — single-agent execution without
+spawning sub-agents. This is efficient for most tasks.
 
-THE ONLY WAY TO SKIP SWARM: User passes -n or --normal explicitly.
+Use `-s`/`--swarm` for multi-agent coordination when the task warrants it.
+Use `-h`/`--hive` for consensus-based coordination on architecture decisions.
+
+POST-TASK NEURAL LEARNING ALWAYS RUNS regardless of execution mode.
+The hooks system collects learnings after every task completion — normal,
+swarm, or hive-mind.
 
 ## COMPREHENSIVE TESTING REQUIREMENT
 
@@ -107,9 +112,9 @@ PR+Done:     Create PR, update issue status, store learnings
 
 | Mode | Description |
 |------|-------------|
-| **SWARM** (default) | Multi-agent via Task tool: researcher, coder, tester, reviewer |
+| **NORMAL** (default) | Single Claude execution. Efficient for most tasks. |
+| **SWARM** (-s) | Multi-agent via Task tool: researcher, coder, tester, reviewer |
 | **HIVE-MIND** (-h) | Consensus-based coordination for architecture decisions |
-| **NORMAL** (-n) | Single Claude, no agent spawning. Only when user explicitly requests. |
 
 ## Phase 1: Research (-r or default first step)
 
@@ -133,7 +138,7 @@ npx flo memory search --query "<issue title keywords>" --namespace patterns
 npx flo memory search --query "<domain keywords>" --namespace guidance
 ```
 
-Or via MCP: `mcp__claude-flow__memory_search`
+Or via MCP: `mcp__moflo__memory_search`
 
 ### 1.4 Read Guidance Docs (ONLY if memory insufficient)
 **Only if memory search returned < 3 relevant results**, read guidance files:
@@ -305,7 +310,7 @@ WORKFLOW STOPS HERE until tests pass. No shortcuts. No exceptions.
 
 ### 4.1 Write and Run Tests
 Write unit, integration, and E2E tests as appropriate for the change type.
-Use the project's existing test runner and patterns.
+Follow the project's established test style, runner, and patterns. If no existing tests or test guidance is present, choose the best options for the project's language and stack, taking compatibility with existing dependencies into account.
 
 ### 4.2 Test Auto-Fix Loop
 If any tests fail, enter the auto-fix loop (max 3 retries OR 10 minutes):
@@ -450,7 +455,7 @@ function extractStories(epicBody) {
 ```javascript
 const args = "$ARGUMENTS".trim().split(/\s+/);
 let workflowMode = "full";    // full, ticket, research
-let execMode = "swarm";       // swarm (default), hive, normal
+let execMode = "normal";      // normal (default), swarm, hive
 let issueNumber = null;
 let titleWords = [];
 
@@ -515,15 +520,15 @@ console.log("SIMPLIFY: /simplify command runs on changed code before PR.");
 
 | Mode | Flag | Description | When to Use |
 |------|------|-------------|-------------|
-| **Swarm** (DEFAULT) | `-s`, `--swarm` | Multi-agent via Task tool | Always, unless explicitly overridden |
+| **Normal** (DEFAULT) | `-n`, `--normal` | Single Claude, no agents | Default for most tasks |
+| **Swarm** | `-s`, `--swarm` | Multi-agent via Task tool | Complex multi-file changes |
 | **Hive-Mind** | `-h`, `--hive` | Consensus-based coordination | Architecture decisions, tradeoffs |
-| **Normal** | `-n`, `--normal` | Single Claude, no agents | User explicitly wants simple mode |
 
 ## Execution Mode Details
 
-### SWARM Mode (Default) - ALWAYS USE UNLESS TOLD OTHERWISE
+### SWARM Mode (-s, --swarm)
 
-You MUST use the Task tool to spawn agents. No exceptions.
+When swarm is requested, you MUST use the Task tool to spawn agents. No exceptions.
 
 **Swarm spawns these agents via Task tool:**
 - `researcher` - Analyzes issue, searches memory, finds patterns
@@ -558,11 +563,12 @@ Use for consensus-based decisions:
 - Approach tradeoffs
 - Design decisions with multiple valid options
 
-### NORMAL Mode (-n, --normal)
+### NORMAL Mode (Default)
 
-**Only when user explicitly requests.** Single Claude execution without agents.
+Single Claude execution without spawning sub-agents.
 - Still uses Task tool for tracking
 - Still creates tasks for visibility
+- Post-task neural learning hooks still fire
 - Just doesn't spawn multiple agents
 
 ---

package/.claude/skills/flo/SKILL.md

@@ -13,7 +13,7 @@ Research, create tickets for, and execute GitHub issues automatically.
 ## Usage
 
 ```
-/flo <issue-number>                   # Full workflow with SWARM (default)
+/flo <issue-number>                   # Full workflow in NORMAL mode (default)
 /flo -t <issue-number>                # Ticket only: research and update ticket, then STOP
 /flo -t <title>                       # Create a NEW ticket with description, acceptance criteria, test cases
 /flo --ticket <issue-number|title>    # Same as -t
@@ -26,8 +26,8 @@ Also available as `/fl` (shorthand alias).
 ### Execution Mode (how work is done)
 
 ```
-/flo 123                              # SWARM mode (default) - multi-agent coordination
-/flo -s 123                           # SWARM mode (explicit)
+/flo 123                              # NORMAL mode (default) - single-agent execution
+/flo -s 123                           # SWARM mode - multi-agent coordination
 /flo --swarm 123                      # Same as -s
 /flo -h 123                           # HIVE-MIND mode - consensus-based coordination
 /flo --hive 123                       # Same as -h
@@ -58,22 +58,27 @@ Also available as `/fl` (shorthand alias).
 ### Combined Examples
 
 ```
-/flo 123                              # Swarm + full workflow (default) - includes ALL tests
+/flo 123                              # Normal + full workflow (default) - includes ALL tests
 /flo 42                               # If #42 is epic, processes stories sequentially
-/flo -t 123                           # Swarm + ticket only (no implementation)
+/flo -s 123                           # Swarm + full workflow (multi-agent coordination)
+/flo -t 123                           # Normal + ticket only (no implementation)
 /flo -h -t 123                        # Hive-mind + ticket only
-/flo -n -r 123                        # Normal + research only
+/flo -s -r 123                        # Swarm + research only
 /flo --swarm --ticket 123             # Explicit swarm + ticket only
-/flo -n 123                           # Normal + full workflow (still runs all tests)
+/flo -n 123                           # Normal (explicit, same as default)
 ```
 
-## SWARM IS MANDATORY BY DEFAULT
+## NORMAL MODE IS THE DEFAULT
 
-Even if a task "looks simple", you MUST use swarm coordination unless
-the user explicitly passes -n/--normal. "Simple" is a trap. Tasks have
-hidden complexity. Swarm catches it.
+By default, /flo runs in NORMAL mode — single-agent execution without
+spawning sub-agents. This is efficient for most tasks.
 
-THE ONLY WAY TO SKIP SWARM: User passes -n or --normal explicitly.
+Use `-s`/`--swarm` for multi-agent coordination when the task warrants it.
+Use `-h`/`--hive` for consensus-based coordination on architecture decisions.
+
+POST-TASK NEURAL LEARNING ALWAYS RUNS regardless of execution mode.
+The hooks system collects learnings after every task completion — normal,
+swarm, or hive-mind.
 
 ## COMPREHENSIVE TESTING REQUIREMENT
 
@@ -107,9 +112,9 @@ PR+Done:     Create PR, update issue status, store learnings
 
 | Mode | Description |
 |------|-------------|
-| **SWARM** (default) | Multi-agent via Task tool: researcher, coder, tester, reviewer |
+| **NORMAL** (default) | Single Claude execution. Efficient for most tasks. |
+| **SWARM** (-s) | Multi-agent via Task tool: researcher, coder, tester, reviewer |
 | **HIVE-MIND** (-h) | Consensus-based coordination for architecture decisions |
-| **NORMAL** (-n) | Single Claude, no agent spawning. Only when user explicitly requests. |
 
 ## Phase 1: Research (-r or default first step)
 
@@ -133,7 +138,7 @@ npx flo memory search --query "<issue title keywords>" --namespace patterns
 npx flo memory search --query "<domain keywords>" --namespace guidance
 ```
 
-Or via MCP: `mcp__claude-flow__memory_search`
+Or via MCP: `mcp__moflo__memory_search`
 
 ### 1.4 Read Guidance Docs (ONLY if memory insufficient)
 **Only if memory search returned < 3 relevant results**, read guidance files:
@@ -305,7 +310,7 @@ WORKFLOW STOPS HERE until tests pass. No shortcuts. No exceptions.
 
 ### 4.1 Write and Run Tests
 Write unit, integration, and E2E tests as appropriate for the change type.
-Use the project's existing test runner and patterns.
+Follow the project's established test style, runner, and patterns. If no existing tests or test guidance is present, choose the best options for the project's language and stack, taking compatibility with existing dependencies into account.
 
 ### 4.2 Test Auto-Fix Loop
 If any tests fail, enter the auto-fix loop (max 3 retries OR 10 minutes):
@@ -450,7 +455,7 @@ function extractStories(epicBody) {
 ```javascript
 const args = "$ARGUMENTS".trim().split(/\s+/);
 let workflowMode = "full";    // full, ticket, research
-let execMode = "swarm";       // swarm (default), hive, normal
+let execMode = "normal";      // normal (default), swarm, hive
 let issueNumber = null;
 let titleWords = [];
 
@@ -515,15 +520,15 @@ console.log("SIMPLIFY: /simplify command runs on changed code before PR.");
 
 | Mode | Flag | Description | When to Use |
 |------|------|-------------|-------------|
-| **Swarm** (DEFAULT) | `-s`, `--swarm` | Multi-agent via Task tool | Always, unless explicitly overridden |
+| **Normal** (DEFAULT) | `-n`, `--normal` | Single Claude, no agents | Default for most tasks |
+| **Swarm** | `-s`, `--swarm` | Multi-agent via Task tool | Complex multi-file changes |
 | **Hive-Mind** | `-h`, `--hive` | Consensus-based coordination | Architecture decisions, tradeoffs |
-| **Normal** | `-n`, `--normal` | Single Claude, no agents | User explicitly wants simple mode |
 
 ## Execution Mode Details
 
-### SWARM Mode (Default) - ALWAYS USE UNLESS TOLD OTHERWISE
+### SWARM Mode (-s, --swarm)
 
-You MUST use the Task tool to spawn agents. No exceptions.
+When swarm is requested, you MUST use the Task tool to spawn agents. No exceptions.
 
 **Swarm spawns these agents via Task tool:**
 - `researcher` - Analyzes issue, searches memory, finds patterns
@@ -558,11 +563,12 @@ Use for consensus-based decisions:
 - Approach tradeoffs
 - Design decisions with multiple valid options
 
-### NORMAL Mode (-n, --normal)
+### NORMAL Mode (Default)
 
-**Only when user explicitly requests.** Single Claude execution without agents.
+Single Claude execution without spawning sub-agents.
 - Still uses Task tool for tracking
 - Still creates tasks for visibility
+- Post-task neural learning hooks still fire
 - Just doesn't spawn multiple agents
 
 ---

package/.claude/skills/github-code-review/SKILL.md

@@ -1027,10 +1027,10 @@ fi
 ```javascript
 [Single Message - Parallel Execution]:
   // Initialize coordination
-  mcp__claude-flow__swarm_init { topology: "hierarchical", maxAgents: 5 }
-  mcp__claude-flow__agent_spawn { type: "reviewer", name: "Senior Reviewer" }
-  mcp__claude-flow__agent_spawn { type: "tester", name: "QA Engineer" }
-  mcp__claude-flow__agent_spawn { type: "coordinator", name: "Merge Coordinator" }
+  mcp__moflo__swarm_init { topology: "hierarchical", maxAgents: 5 }
+  mcp__moflo__agent_spawn { type: "reviewer", name: "Senior Reviewer" }
+  mcp__moflo__agent_spawn { type: "tester", name: "QA Engineer" }
+  mcp__moflo__agent_spawn { type: "coordinator", name: "Merge Coordinator" }
 
   // Create and manage PR using gh CLI
   Bash("gh pr create --title 'Feature: Add authentication' --base main")

package/.claude/skills/github-multi-repo/SKILL.md

@@ -90,7 +90,7 @@ const DEPS = Bash(`gh repo list my-organization --json name | \
   done | jq -s '.'`)
 
 // Initialize swarm with discovered repositories
-mcp__claude-flow__swarm_init({
+mcp__moflo__swarm_init({
   topology: "hierarchical",
   maxAgents: 8,
   metadata: { repos: REPOS, dependencies: DEPS }
@@ -145,7 +145,7 @@ mcp__claude-flow__swarm_init({
 // Synchronize package dependencies and versions
 [Complete Package Sync]:
   // Initialize sync swarm
-  mcp__claude-flow__swarm_init({ topology: "mesh", maxAgents: 5 })
+  mcp__moflo__swarm_init({ topology: "mesh", maxAgents: 5 })
 
   // Spawn sync agents
   Task("Sync Coordinator", "Coordinate version alignment", "coordinator")
@@ -169,7 +169,7 @@ mcp__claude-flow__swarm_init({
     -f content="$(cat aligned-package.json | base64)"`)
 
   // Store sync state
-  mcp__claude-flow__memory_usage({
+  mcp__moflo__memory_usage({
     action: "store",
     key: "sync/packages/status",
     value: {
@@ -196,7 +196,7 @@ mcp__claude-flow__swarm_init({
     -f content="$(cat /tmp/claude-source.md | base64)"`)
 
   // Track sync status
-  mcp__claude-flow__memory_usage({
+  mcp__moflo__memory_usage({
     action: "store",
     key: "sync/documentation/status",
     value: { status: "synchronized", files: ["CLAUDE.md"] }
@@ -246,7 +246,7 @@ mcp__claude-flow__swarm_init({
 // Analyze and optimize repository structure
 [Architecture Analysis]:
   // Initialize architecture swarm
-  mcp__claude-flow__swarm_init({ topology: "hierarchical", maxAgents: 6 })
+  mcp__moflo__swarm_init({ topology: "hierarchical", maxAgents: 6 })
 
   // Spawn architecture agents
   Task("Senior Architect", "Analyze repository structure", "architect")
@@ -266,7 +266,7 @@ mcp__claude-flow__swarm_init({
     --order desc`)
 
   // Store analysis results
-  mcp__claude-flow__memory_usage({
+  mcp__moflo__memory_usage({
     action: "store",
     key: "architecture/analysis/results",
     value: {
@@ -406,7 +406,7 @@ Part of #$TRACKING_ISSUE"
 // Coordinate large-scale refactoring
 [Cross-Repo Refactoring]:
   // Initialize refactoring swarm
-  mcp__claude-flow__swarm_init({ topology: "mesh", maxAgents: 8 })
+  mcp__moflo__swarm_init({ topology: "mesh", maxAgents: 8 })
 
   // Spawn specialized agents
   Task("Refactoring Coordinator", "Coordinate refactoring across repos", "coordinator")
@@ -416,7 +416,7 @@ Part of #$TRACKING_ISSUE"
   Task("Integration Tester", "Validate refactored code", "tester")
 
   // Execute refactoring
-  mcp__claude-flow__task_orchestrate({
+  mcp__moflo__task_orchestrate({
     task: "Rename OldAPI to NewAPI across all repositories",
     strategy: "sequential",
     priority: "high"

package/.claude/skills/github-project-management/SKILL.md

@@ -20,7 +20,7 @@ prerequisites:
   - Repository access permissions
 tools_required:
   - mcp__github__*
-  - mcp__claude-flow__*
+  - mcp__moflo__*
   - Bash
   - Read
   - Write
@@ -79,10 +79,10 @@ npx ruv-swarm github board-init \
 
 ```javascript
 // Initialize issue management swarm
-mcp__claude-flow__swarm_init { topology: "star", maxAgents: 3 }
-mcp__claude-flow__agent_spawn { type: "coordinator", name: "Issue Coordinator" }
-mcp__claude-flow__agent_spawn { type: "researcher", name: "Requirements Analyst" }
-mcp__claude-flow__agent_spawn { type: "coder", name: "Implementation Planner" }
+mcp__moflo__swarm_init { topology: "star", maxAgents: 3 }
+mcp__moflo__agent_spawn { type: "coordinator", name: "Issue Coordinator" }
+mcp__moflo__agent_spawn { type: "researcher", name: "Requirements Analyst" }
+mcp__moflo__agent_spawn { type: "coder", name: "Implementation Planner" }
 
 // Create comprehensive issue
 mcp__github__create_issue {
@@ -107,7 +107,7 @@ mcp__github__create_issue {
 }
 
 // Set up automated tracking
-mcp__claude-flow__task_orchestrate {
+mcp__moflo__task_orchestrate {
   task: "Monitor and coordinate issue progress with automated updates",
   strategy: "adaptive",
   priority: "medium"

package/.claude/skills/github-release-management/SKILL.md

@@ -150,19 +150,19 @@ gh release create $(npm pkg get version) \
 ```javascript
 // Set up coordinated release team
 [Single Message - Swarm Initialization]:
-  mcp__claude-flow__swarm_init {
+  mcp__moflo__swarm_init {
     topology: "hierarchical",
     maxAgents: 6,
     strategy: "balanced"
   }
 
   // Spawn specialized agents
-  mcp__claude-flow__agent_spawn { type: "coordinator", name: "Release Director" }
-  mcp__claude-flow__agent_spawn { type: "coder", name: "Version Manager" }
-  mcp__claude-flow__agent_spawn { type: "tester", name: "QA Engineer" }
-  mcp__claude-flow__agent_spawn { type: "reviewer", name: "Release Reviewer" }
-  mcp__claude-flow__agent_spawn { type: "analyst", name: "Deployment Analyst" }
-  mcp__claude-flow__agent_spawn { type: "researcher", name: "Compatibility Checker" }
+  mcp__moflo__agent_spawn { type: "coordinator", name: "Release Director" }
+  mcp__moflo__agent_spawn { type: "coder", name: "Version Manager" }
+  mcp__moflo__agent_spawn { type: "tester", name: "QA Engineer" }
+  mcp__moflo__agent_spawn { type: "reviewer", name: "Release Reviewer" }
+  mcp__moflo__agent_spawn { type: "analyst", name: "Deployment Analyst" }
+  mcp__moflo__agent_spawn { type: "researcher", name: "Compatibility Checker" }
 ```
 
 #### Coordinated Release Workflow
@@ -172,7 +172,7 @@ gh release create $(npm pkg get version) \
   Bash("gh api repos/:owner/:repo/git/refs --method POST -f ref='refs/heads/release/v2.0.0' -f sha=$(gh api repos/:owner/:repo/git/refs/heads/main --jq '.object.sha')")
 
   // Orchestrate release preparation
-  mcp__claude-flow__task_orchestrate {
+  mcp__moflo__task_orchestrate {
     task: "Prepare release v2.0.0 with comprehensive testing and validation",
     strategy: "sequential",
     priority: "critical",
@@ -204,7 +204,7 @@ gh release create $(npm pkg get version) \
   ]}
 
   // Store release state
-  mcp__claude-flow__memory_usage {
+  mcp__moflo__memory_usage {
     action: "store",
     key: "release/v2.0.0/status",
     value: JSON.stringify({
@@ -309,7 +309,7 @@ npx claude-flow github release-deploy \
 ```javascript
 [Single Message - Multi-Package Release]:
   // Initialize mesh topology for cross-package coordination
-  mcp__claude-flow__swarm_init { topology: "mesh", maxAgents: 8 }
+  mcp__moflo__swarm_init { topology: "mesh", maxAgents: 8 }
 
   // Spawn package-specific agents
   Task("Package A Manager", "Coordinate claude-flow package release v1.0.72", "coder")
@@ -394,7 +394,7 @@ npx claude-flow github multi-release \
 ```javascript
 [Single Message - Cross-Repo Release]:
   // Initialize star topology for centralized coordination
-  mcp__claude-flow__swarm_init { topology: "star", maxAgents: 6 }
+  mcp__moflo__swarm_init { topology: "star", maxAgents: 6 }
 
   // Spawn repo-specific coordinators
   Task("Frontend Release", "Release frontend v2.0.0 with API compatibility", "coordinator")
@@ -408,7 +408,7 @@ npx claude-flow github multi-release \
   Bash("gh api repos/org/cli/dispatches --method POST -f event_type='release' -F client_payload[version]=v1.5.0")
 
   // Monitor all releases
-  mcp__claude-flow__swarm_monitor { interval: 5, duration: 300 }
+  mcp__moflo__swarm_monitor { interval: 5, duration: 300 }
 ```
 
 ### Hotfix Emergency Procedures

package/.claude/skills/github-workflow-automation/SKILL.md

@@ -537,19 +537,19 @@ run().catch(error => core.setFailed(error.message));
 #### Initialize GitHub Swarm
 ```javascript
 // Step 1: Initialize swarm coordination
-mcp__claude-flow__swarm_init {
+mcp__moflo__swarm_init {
   topology: "hierarchical",
   maxAgents: 8
 }
 
 // Step 2: Spawn specialized agents
-mcp__claude-flow__agent_spawn { type: "coordinator", name: "GitHub Coordinator" }
-mcp__claude-flow__agent_spawn { type: "reviewer", name: "Code Reviewer" }
-mcp__claude-flow__agent_spawn { type: "tester", name: "QA Agent" }
-mcp__claude-flow__agent_spawn { type: "analyst", name: "Security Analyst" }
+mcp__moflo__agent_spawn { type: "coordinator", name: "GitHub Coordinator" }
+mcp__moflo__agent_spawn { type: "reviewer", name: "Code Reviewer" }
+mcp__moflo__agent_spawn { type: "tester", name: "QA Agent" }
+mcp__moflo__agent_spawn { type: "analyst", name: "Security Analyst" }
 
 // Step 3: Orchestrate GitHub workflow
-mcp__claude-flow__task_orchestrate {
+mcp__moflo__task_orchestrate {
   task: "Complete PR review and merge workflow",
   strategy: "parallel",
   priority: "high"