claude-flow-novice 2.14.21 → 2.14.23

package/claude-assets/agents/cfn-dev-team/CLAUDE.md

@@ -68,9 +68,7 @@ touch .claude/agents/custom/my-first-agent.md
 ```markdown
 ---
 name: my-first-agent
-description: |
-  MUST BE USED when performing [your specific task].
-  Keywords - [relevant, keywords, for, your, task]
+description: MUST BE USED when performing [your specific task]. Keywords - [relevant, keywords, for, your, task]
 tools: [Read, Write, Edit, Bash, TodoWrite]
 model: haiku
 type: specialist
@@ -156,9 +154,7 @@ Every agent has three main parts:
 ```yaml
 ---
 name: agent-name                    # Unique identifier (lowercase, hyphens)
-description: |                      # What the agent does & when to use it
-  MUST BE USED when [use case].
-  Keywords - [searchable, terms]
+description: MUST BE USED when [use case]. Keywords - [searchable, terms]  # Single-line for token efficiency
 tools: [Read, Write, Edit, Bash]    # Available tools (Bash includes Redis CLI)
 model: haiku                        # haiku | sonnet | opus
 type: specialist                    # specialist | coordinator | swarm
@@ -239,10 +235,7 @@ touch .claude/agents/custom/migration-reviewer.md
 ```markdown
 ---
 name: migration-reviewer
-description: |
-  MUST BE USED when reviewing database migrations for security and performance.
-  Use PROACTIVELY for SQL schema changes, migrations, database updates.
-  Keywords - database, migration, schema, SQL, security, performance
+description: MUST BE USED when reviewing database migrations for security and performance. Use PROACTIVELY for SQL schema changes, migrations, database updates. Keywords - database, migration, schema, SQL, security, performance
 tools: [Read, Grep, TodoWrite]
 model: haiku
 type: specialist
@@ -319,9 +312,7 @@ npx claude-flow-novice agent-spawn migration-reviewer \
 ```markdown
 ---
 name: my-implementer
-description: |
-  MUST BE USED when implementing [specific type of code].
-  Keywords - implement, build, create, develop
+description: MUST BE USED when implementing [specific type of code]. Keywords - implement, build, create, develop
 tools: [Read, Write, Edit, Bash, TodoWrite]
 model: haiku
 type: specialist
@@ -357,9 +348,7 @@ You implement [specific type of code] following best practices.
 ```markdown
 ---
 name: my-reviewer
-description: |
-  MUST BE USED for reviewing [specific type of code].
-  Keywords - review, validate, quality, security
+description: MUST BE USED for reviewing [specific type of code]. Keywords - review, validate, quality, security
 tools: [Read, Grep, TodoWrite]
 model: haiku
 type: validator
@@ -407,10 +396,7 @@ Provide confidence score and categorized feedback:
 ```markdown
 ---
 name: my-specialist
-description: |
-  MUST BE USED when [specialized task].
-  Use PROACTIVELY for [scenarios].
-  Keywords - [domain, specific, keywords]
+description: MUST BE USED when [specialized task]. Use PROACTIVELY for [scenarios]. Keywords - [domain, specific, keywords]
 tools: [Read, Write, Edit, Bash, Grep, Glob, TodoWrite]
 model: sonnet
 type: specialist
@@ -744,10 +730,7 @@ Here's a real-world example - a Terraform reviewer agent:
 ```markdown
 ---
 name: terraform-reviewer
-description: |
-  MUST BE USED when reviewing Terraform infrastructure code.
-  Use PROACTIVELY for IaC review, security validation, cost optimization.
-  Keywords - terraform, infrastructure, IaC, cloud, AWS, security
+description: MUST BE USED when reviewing Terraform infrastructure code. Use PROACTIVELY for IaC review, security validation, cost optimization. Keywords - terraform, infrastructure, IaC, cloud, AWS, security
 tools: [Read, Grep, Glob, TodoWrite]
 model: sonnet
 type: specialist
@@ -912,8 +895,7 @@ acl_level: 3
 ```yaml
 ---
 name: agent-name
-description: |
-  What the agent does and when to use it
+description: What the agent does and when to use it
 tools: [Read, Write]
 model: haiku
 ---
@@ -923,9 +905,7 @@ model: haiku
 ```yaml
 ---
 name: agent-name
-description: |
-  MUST BE USED when [use case]
-  Keywords - [keywords]
+description: MUST BE USED when [use case] Keywords - [keywords]  # Single-line for token efficiency
 tools: [Read, Write, Edit, TodoWrite]
 model: haiku
 type: specialist
@@ -1009,14 +989,35 @@ Now that you understand agent creation, explore:
 
 ---
 
-## CFN Loop Completion Protocol (CRITICAL UPDATE)
+## CFN Loop Completion Protocol (MODE-SPECIFIC)
 
-**Date:** 2025-10-30
-**Status:** Waiting Mode Deprecated
+**Date:** 2025-11-06
+**Status:** Memory Leak Fix Applied - ANTI-023 Resolution
 
-### Current Standard Protocol
+### ⚠️ CRITICAL: Mode-Specific Completion Required
 
-All agents participating in CFN Loop workflows MUST follow this protocol:
+Agents MUST use different completion protocols based on how they were spawned:
+
+### Task Mode (Spawned via Task() tool in Main Chat)
+
+**Simply complete your work and return structured output.**
+
+```json
+{
+  "confidence": 0.85,
+  "status": "COMPLETE|NEEDS_WORK",
+  "summary": "Brief summary of work completed",
+  "deliverables": ["file1.ts", "file2.test.ts"]
+}
+```
+
+**❌ DO NOT:**
+- Use Redis commands (redis-cli)
+- Execute bash scripts for completion
+- Signal completion via CLI tools
+- **Main Chat receives your output automatically**
+
+### CLI Mode (Spawned via `npx claude-flow-novice agent-spawn`)
 
 **Step 1: Complete Work**
 Execute assigned task (implementation, validation, review, etc.)
@@ -1035,52 +1036,26 @@ redis-cli lpush "swarm:${TASK_ID}:${AGENT_ID}:done" "complete"
   --iteration 1
 ```
 
-**After reporting, exit cleanly. Do NOT enter waiting mode.**
-
-### Why This Matters
-
-- Orchestrator collects confidence/consensus scores from Redis
-- Enables adaptive agent specialization for next iteration
-- Prevents orchestrator blocking on `wait $PID`
-- Coordinator spawns appropriate specialist based on feedback type
+### How to Tell Which Mode You're In
 
-### Deprecated Patterns (DO NOT USE)
+- **Task Mode**: You were spawned via `Task("agent-name", "...")` in Main Chat
+- **CLI Mode**: You were spawned via `npx claude-flow-novice agent-spawn ...` command
 
-❌ **Step 4: Enter Waiting Mode** (REMOVED)
-```bash
-# DEPRECATED - This will cause errors
-./.claude/skills/cfn-redis-coordination/invoke-waiting-mode.sh enter \
-  --task-id "$TASK_ID" --agent-id "$AGENT_ID"
-```
-
-❌ **Wake Agents for Iteration** (REMOVED)
-```bash
-# DEPRECATED - Spawn fresh agents instead
-./.claude/skills/cfn-redis-coordination/invoke-waiting-mode.sh wake \
-  --task-id "$TASK_ID" --agent-id "$AGENT_ID"
-```
-
-### Correct Iteration Pattern
+### Why This Matters
 
-**Coordinators:** Spawn fresh agents for each iteration:
-```bash
-# Iteration N+1
-for agent in "${agents[@]}"; do
-  npx claude-flow-novice agent-spawn "$agent" \
-    --task-id "$TASK_ID" \
-    --context "Iteration $iteration: Address feedback [...]"
-done
-```
+- **Task Mode**: Main Chat handles everything, just return results
+- **CLI Mode**: Coordinator needs Redis signals to collect confidence scores
+- **Mixed protocols cause memory leaks** (ANTI-023 pattern)
 
 ### Related Documentation
 
-- **Bug Fix Details:** `docs/BUG_WAITING_MODE_FIX.md`
+- **Memory Leak Fix:** `docs/bugs/BUG_MEMORY_LEAK_VALIDATOR_FIX.md`
 - **Agent Lifecycle:** `.claude/agents/AGENT_LIFECYCLE.md`
-- **Redis Coordination:** `.claude/skills/cfn-redis-coordination/SKILL.md`
+- **Main Documentation:** `CLAUDE.md:333-357` (Mode-specific protocols)
 
 ---
 
-**Document Version:** 4.1.0 (Waiting Mode Deprecation Update)
-**Last Updated:** 2025-10-30
+**Document Version:** 4.2.0 (Memory Leak Fix - Mode-Specific Protocols)
+**Last Updated:** 2025-11-06
 **Maintained By:** Claude Flow Novice Team
 **Feedback:** We'd love to hear how you're using agents! Share your creations.

package/claude-assets/agents/cfn-dev-team/analysts/root-cause-analyst.md

@@ -1,9 +1,6 @@
 ---
 name: root-cause-analyst
-description: |
-  MUST BE USED when investigating technical issues, bugs, system failures to identify true root causes.
-  Use PROACTIVELY for deep technical investigation, error analysis, failure diagnosis, debugging complex issues.
-  Keywords - root cause, investigation, bug analysis, failure diagnosis, debugging, error tracing, issue investigation
+description: MUST BE USED when investigating technical issues, bugs, system failures to identify true root causes. Use PROACTIVELY for deep technical investigation, error analysis, failure diagnosis, debugging complex issues. Keywords - root cause, investigation, bug analysis, failure diagnosis, debugging, error tracing, issue investigation
 tools: [Bash, Grep, Glob]
 model: sonnet
 type: specialist

package/claude-assets/agents/cfn-dev-team/architecture/goal-planner.md

@@ -1,9 +1,6 @@
 ---
 name: goal-planner
-description: |
-  MUST BE USED for goal-oriented action planning with A* search algorithms.
-  Use PROACTIVELY for multi-phase planning, goal decomposition, strategic decision-making.
-  Keywords - GOAP, A* search, planning, goal decomposition, adaptive replanning
+description: MUST BE USED for goal-oriented action planning with A* search algorithms. Use PROACTIVELY for multi-phase planning, goal decomposition, strategic decision-making. Keywords - GOAP, A* search, planning, goal decomposition, adaptive replanning
 tools: [Read, Write, Edit, Bash, TodoWrite, mcp__claude-flow-novice__memory_usage]
 model: haiku
 color: purple

package/claude-assets/agents/cfn-dev-team/architecture/planner.md

@@ -1,9 +1,6 @@
 ---
 name: planner
-description: |
-  MUST BE USED for generic task breakdown, basic coordination, milestone planning.
-  FALLBACK agent when no specialized planner (goal-planner, architect, project managers) exists.
-  Keywords - task decomposition, coordination, planning, milestone tracking
+description: MUST BE USED for generic task breakdown, basic coordination, milestone planning. FALLBACK agent when no specialized planner (goal-planner, architect, project managers) exists. Keywords - task decomposition, coordination, planning, milestone tracking
 tools: [TodoWrite, Read, Write, Edit, Bash, Glob, Grep]
 model: haiku
 type: coordinator

package/claude-assets/agents/cfn-dev-team/architecture/system-architect.md

@@ -1,9 +1,6 @@
 ---
 name: system-architect
-description: |
-  MUST BE USED when designing enterprise-grade system architecture, providing technical leadership for distributed systems, microservices, cloud-native solutions.
-  Use PROACTIVELY for architectural design, technical strategy, infrastructure planning.
-  Keywords - enterprise architecture, system design, technical leadership, architectural patterns
+description: MUST BE USED when designing enterprise-grade system architecture, providing technical leadership for distributed systems, microservices, cloud-native solutions. Use PROACTIVELY for architectural design, technical strategy, infrastructure planning. Keywords - enterprise architecture, system design, technical leadership, architectural patterns
 tools: [Read, Write, Edit, Bash, Grep, Glob, TodoWrite]
 model: haiku
 color: seagreen