@techwavedev/agi-agent-kit 1.3.5 → 1.4.2

package/templates/skills/extended/architecture/projection-patterns/SKILL.md

@@ -40,15 +40,18 @@ Comprehensive guide to building projections and read models for event-sourced sy
 > **Adapted for [@techwavedev/agi-agent-kit](https://www.npmjs.com/package/@techwavedev/agi-agent-kit)**
 > Original source: [antigravity-awesome-skills](https://github.com/sickn33/antigravity-awesome-skills)
 
-### Qdrant Memory Integration
+### Hybrid Memory Integration (Qdrant + BM25)
 
 Before executing complex tasks with this skill:
 ```bash
 python3 execution/memory_manager.py auto --query "<task summary>"
 ```
+
+**Decision Tree:**
 - **Cache hit?** Use cached response directly — no need to re-process.
 - **Memory match?** Inject `context_chunks` into your reasoning.
 - **No match?** Proceed normally, then store results:
+
 ```bash
 python3 execution/memory_manager.py store \
   --content "Description of what was decided/solved" \
@@ -56,14 +59,16 @@ python3 execution/memory_manager.py store \
   --tags projection-patterns <relevant-tags>
 ```
 
+> **Note:** Storing automatically updates both Vector (Qdrant) and Keyword (BM25) indices.
+
 ### Agent Team Collaboration
 
-- This skill can be invoked by the `orchestrator` agent via intelligent routing.
-- In **Agent Teams mode**, results are shared via Qdrant shared memory for cross-agent context.
-- In **Subagent mode**, this skill runs in isolation with its own memory namespace.
+- **Strategy**: This skill communicates via the shared memory system.
+- **Orchestration**: Invoked by `orchestrator` via intelligent routing.
+- **Context Sharing**: Always read previous agent outputs from memory before starting.
 
 ### Local LLM Support
 
 When available, use local Ollama models for embedding and lightweight inference:
 - Embeddings: `nomic-embed-text` via Qdrant memory system
-- Lightweight analysis: Local models reduce API costs for repetitive patterns
+- Lightweight analysis: Local models reduce API costs for repetitive patterns

package/templates/skills/extended/architecture/rust-async-patterns/SKILL.md

@@ -40,15 +40,18 @@ Production patterns for async Rust programming with Tokio runtime, including tas
 > **Adapted for [@techwavedev/agi-agent-kit](https://www.npmjs.com/package/@techwavedev/agi-agent-kit)**
 > Original source: [antigravity-awesome-skills](https://github.com/sickn33/antigravity-awesome-skills)
 
-### Qdrant Memory Integration
+### Hybrid Memory Integration (Qdrant + BM25)
 
 Before executing complex tasks with this skill:
 ```bash
 python3 execution/memory_manager.py auto --query "<task summary>"
 ```
+
+**Decision Tree:**
 - **Cache hit?** Use cached response directly — no need to re-process.
 - **Memory match?** Inject `context_chunks` into your reasoning.
 - **No match?** Proceed normally, then store results:
+
 ```bash
 python3 execution/memory_manager.py store \
   --content "Description of what was decided/solved" \
@@ -56,14 +59,16 @@ python3 execution/memory_manager.py store \
   --tags rust-async-patterns <relevant-tags>
 ```
 
+> **Note:** Storing automatically updates both Vector (Qdrant) and Keyword (BM25) indices.
+
 ### Agent Team Collaboration
 
-- This skill can be invoked by the `orchestrator` agent via intelligent routing.
-- In **Agent Teams mode**, results are shared via Qdrant shared memory for cross-agent context.
-- In **Subagent mode**, this skill runs in isolation with its own memory namespace.
+- **Strategy**: This skill communicates via the shared memory system.
+- **Orchestration**: Invoked by `orchestrator` via intelligent routing.
+- **Context Sharing**: Always read previous agent outputs from memory before starting.
 
 ### Local LLM Support
 
 When available, use local Ollama models for embedding and lightweight inference:
 - Embeddings: `nomic-embed-text` via Qdrant memory system
-- Lightweight analysis: Local models reduce API costs for repetitive patterns
+- Lightweight analysis: Local models reduce API costs for repetitive patterns

package/templates/skills/extended/architecture/software-architecture/SKILL.md

@@ -82,15 +82,18 @@ This skill provides guidance for quality focused software development and archit
 > **Adapted for [@techwavedev/agi-agent-kit](https://www.npmjs.com/package/@techwavedev/agi-agent-kit)**
 > Original source: [antigravity-awesome-skills](https://github.com/sickn33/antigravity-awesome-skills)
 
-### Qdrant Memory Integration
+### Hybrid Memory Integration (Qdrant + BM25)
 
 Before executing complex tasks with this skill:
 ```bash
 python3 execution/memory_manager.py auto --query "<task summary>"
 ```
+
+**Decision Tree:**
 - **Cache hit?** Use cached response directly — no need to re-process.
 - **Memory match?** Inject `context_chunks` into your reasoning.
 - **No match?** Proceed normally, then store results:
+
 ```bash
 python3 execution/memory_manager.py store \
   --content "Description of what was decided/solved" \
@@ -98,14 +101,16 @@ python3 execution/memory_manager.py store \
   --tags software-architecture <relevant-tags>
 ```
 
+> **Note:** Storing automatically updates both Vector (Qdrant) and Keyword (BM25) indices.
+
 ### Agent Team Collaboration
 
-- This skill can be invoked by the `orchestrator` agent via intelligent routing.
-- In **Agent Teams mode**, results are shared via Qdrant shared memory for cross-agent context.
-- In **Subagent mode**, this skill runs in isolation with its own memory namespace.
+- **Strategy**: This skill communicates via the shared memory system.
+- **Orchestration**: Invoked by `orchestrator` via intelligent routing.
+- **Context Sharing**: Always read previous agent outputs from memory before starting.
 
 ### Local LLM Support
 
 When available, use local Ollama models for embedding and lightweight inference:
 - Embeddings: `nomic-embed-text` via Qdrant memory system
-- Lightweight analysis: Local models reduce API costs for repetitive patterns
+- Lightweight analysis: Local models reduce API costs for repetitive patterns

package/templates/skills/extended/architecture/stride-analysis-patterns/SKILL.md

@@ -40,15 +40,18 @@ Systematic threat identification using the STRIDE methodology.
 > **Adapted for [@techwavedev/agi-agent-kit](https://www.npmjs.com/package/@techwavedev/agi-agent-kit)**
 > Original source: [antigravity-awesome-skills](https://github.com/sickn33/antigravity-awesome-skills)
 
-### Qdrant Memory Integration
+### Hybrid Memory Integration (Qdrant + BM25)
 
 Before executing complex tasks with this skill:
 ```bash
 python3 execution/memory_manager.py auto --query "<task summary>"
 ```
+
+**Decision Tree:**
 - **Cache hit?** Use cached response directly — no need to re-process.
 - **Memory match?** Inject `context_chunks` into your reasoning.
 - **No match?** Proceed normally, then store results:
+
 ```bash
 python3 execution/memory_manager.py store \
   --content "Description of what was decided/solved" \
@@ -56,14 +59,16 @@ python3 execution/memory_manager.py store \
   --tags stride-analysis-patterns <relevant-tags>
 ```
 
+> **Note:** Storing automatically updates both Vector (Qdrant) and Keyword (BM25) indices.
+
 ### Agent Team Collaboration
 
-- This skill can be invoked by the `orchestrator` agent via intelligent routing.
-- In **Agent Teams mode**, results are shared via Qdrant shared memory for cross-agent context.
-- In **Subagent mode**, this skill runs in isolation with its own memory namespace.
+- **Strategy**: This skill communicates via the shared memory system.
+- **Orchestration**: Invoked by `orchestrator` via intelligent routing.
+- **Context Sharing**: Always read previous agent outputs from memory before starting.
 
 ### Local LLM Support
 
 When available, use local Ollama models for embedding and lightweight inference:
 - Embeddings: `nomic-embed-text` via Qdrant memory system
-- Lightweight analysis: Local models reduce API costs for repetitive patterns
+- Lightweight analysis: Local models reduce API costs for repetitive patterns

package/templates/skills/extended/architecture/team-composition-analysis/SKILL.md

@@ -420,15 +420,18 @@ For detailed compensation benchmarks and hiring plan templates, see `references/
 > **Adapted for [@techwavedev/agi-agent-kit](https://www.npmjs.com/package/@techwavedev/agi-agent-kit)**
 > Original source: [antigravity-awesome-skills](https://github.com/sickn33/antigravity-awesome-skills)
 
-### Qdrant Memory Integration
+### Hybrid Memory Integration (Qdrant + BM25)
 
 Before executing complex tasks with this skill:
 ```bash
 python3 execution/memory_manager.py auto --query "<task summary>"
 ```
+
+**Decision Tree:**
 - **Cache hit?** Use cached response directly — no need to re-process.
 - **Memory match?** Inject `context_chunks` into your reasoning.
 - **No match?** Proceed normally, then store results:
+
 ```bash
 python3 execution/memory_manager.py store \
   --content "Description of what was decided/solved" \
@@ -436,14 +439,16 @@ python3 execution/memory_manager.py store \
   --tags team-composition-analysis <relevant-tags>
 ```
 
+> **Note:** Storing automatically updates both Vector (Qdrant) and Keyword (BM25) indices.
+
 ### Agent Team Collaboration
 
-- This skill can be invoked by the `orchestrator` agent via intelligent routing.
-- In **Agent Teams mode**, results are shared via Qdrant shared memory for cross-agent context.
-- In **Subagent mode**, this skill runs in isolation with its own memory namespace.
+- **Strategy**: This skill communicates via the shared memory system.
+- **Orchestration**: Invoked by `orchestrator` via intelligent routing.
+- **Context Sharing**: Always read previous agent outputs from memory before starting.
 
 ### Local LLM Support
 
 When available, use local Ollama models for embedding and lightweight inference:
 - Embeddings: `nomic-embed-text` via Qdrant memory system
-- Lightweight analysis: Local models reduce API costs for repetitive patterns
+- Lightweight analysis: Local models reduce API costs for repetitive patterns

package/templates/skills/extended/architecture/workflow-patterns/SKILL.md

@@ -45,15 +45,18 @@ Guide for implementing tasks using Conductor's TDD workflow, managing phase chec
 > **Adapted for [@techwavedev/agi-agent-kit](https://www.npmjs.com/package/@techwavedev/agi-agent-kit)**
 > Original source: [antigravity-awesome-skills](https://github.com/sickn33/antigravity-awesome-skills)
 
-### Qdrant Memory Integration
+### Hybrid Memory Integration (Qdrant + BM25)
 
 Before executing complex tasks with this skill:
 ```bash
 python3 execution/memory_manager.py auto --query "<task summary>"
 ```
+
+**Decision Tree:**
 - **Cache hit?** Use cached response directly — no need to re-process.
 - **Memory match?** Inject `context_chunks` into your reasoning.
 - **No match?** Proceed normally, then store results:
+
 ```bash
 python3 execution/memory_manager.py store \
   --content "Description of what was decided/solved" \
@@ -61,14 +64,16 @@ python3 execution/memory_manager.py store \
   --tags workflow-patterns <relevant-tags>
 ```
 
+> **Note:** Storing automatically updates both Vector (Qdrant) and Keyword (BM25) indices.
+
 ### Agent Team Collaboration
 
-- This skill can be invoked by the `orchestrator` agent via intelligent routing.
-- In **Agent Teams mode**, results are shared via Qdrant shared memory for cross-agent context.
-- In **Subagent mode**, this skill runs in isolation with its own memory namespace.
+- **Strategy**: This skill communicates via the shared memory system.
+- **Orchestration**: Invoked by `orchestrator` via intelligent routing.
+- **Context Sharing**: Always read previous agent outputs from memory before starting.
 
 ### Local LLM Support
 
 When available, use local Ollama models for embedding and lightweight inference:
 - Embeddings: `nomic-embed-text` via Qdrant memory system
-- Lightweight analysis: Local models reduce API costs for repetitive patterns
+- Lightweight analysis: Local models reduce API costs for repetitive patterns

package/templates/skills/extended/architecture/writing-plans/SKILL.md

@@ -123,15 +123,18 @@ After saving the plan, offer execution choice:
 > **Adapted for [@techwavedev/agi-agent-kit](https://www.npmjs.com/package/@techwavedev/agi-agent-kit)**
 > Original source: [antigravity-awesome-skills](https://github.com/sickn33/antigravity-awesome-skills)
 
-### Qdrant Memory Integration
+### Hybrid Memory Integration (Qdrant + BM25)
 
 Before executing complex tasks with this skill:
 ```bash
 python3 execution/memory_manager.py auto --query "<task summary>"
 ```
+
+**Decision Tree:**
 - **Cache hit?** Use cached response directly — no need to re-process.
 - **Memory match?** Inject `context_chunks` into your reasoning.
 - **No match?** Proceed normally, then store results:
+
 ```bash
 python3 execution/memory_manager.py store \
   --content "Description of what was decided/solved" \
@@ -139,14 +142,16 @@ python3 execution/memory_manager.py store \
   --tags writing-plans <relevant-tags>
 ```
 
+> **Note:** Storing automatically updates both Vector (Qdrant) and Keyword (BM25) indices.
+
 ### Agent Team Collaboration
 
-- This skill can be invoked by the `orchestrator` agent via intelligent routing.
-- In **Agent Teams mode**, results are shared via Qdrant shared memory for cross-agent context.
-- In **Subagent mode**, this skill runs in isolation with its own memory namespace.
+- **Strategy**: This skill communicates via the shared memory system.
+- **Orchestration**: Invoked by `orchestrator` via intelligent routing.
+- **Context Sharing**: Always read previous agent outputs from memory before starting.
 
 ### Local LLM Support
 
 When available, use local Ollama models for embedding and lightweight inference:
 - Embeddings: `nomic-embed-text` via Qdrant memory system
-- Lightweight analysis: Local models reduce API costs for repetitive patterns
+- Lightweight analysis: Local models reduce API costs for repetitive patterns

package/templates/skills/extended/backend/anti-reversing-techniques/SKILL.md

@@ -49,15 +49,18 @@ Refer to `resources/implementation-playbook.md` for detailed techniques and exam
 > **Adapted for [@techwavedev/agi-agent-kit](https://www.npmjs.com/package/@techwavedev/agi-agent-kit)**
 > Original source: [antigravity-awesome-skills](https://github.com/sickn33/antigravity-awesome-skills)
 
-### Qdrant Memory Integration
+### Hybrid Memory Integration (Qdrant + BM25)
 
 Before executing complex tasks with this skill:
 ```bash
 python3 execution/memory_manager.py auto --query "<task summary>"
 ```
+
+**Decision Tree:**
 - **Cache hit?** Use cached response directly — no need to re-process.
 - **Memory match?** Inject `context_chunks` into your reasoning.
 - **No match?** Proceed normally, then store results:
+
 ```bash
 python3 execution/memory_manager.py store \
   --content "Description of what was decided/solved" \
@@ -65,14 +68,16 @@ python3 execution/memory_manager.py store \
   --tags anti-reversing-techniques <relevant-tags>
 ```
 
+> **Note:** Storing automatically updates both Vector (Qdrant) and Keyword (BM25) indices.
+
 ### Agent Team Collaboration
 
-- This skill can be invoked by the `orchestrator` agent via intelligent routing.
-- In **Agent Teams mode**, results are shared via Qdrant shared memory for cross-agent context.
-- In **Subagent mode**, this skill runs in isolation with its own memory namespace.
+- **Strategy**: This skill communicates via the shared memory system.
+- **Orchestration**: Invoked by `orchestrator` via intelligent routing.
+- **Context Sharing**: Always read previous agent outputs from memory before starting.
 
 ### Local LLM Support
 
 When available, use local Ollama models for embedding and lightweight inference:
 - Embeddings: `nomic-embed-text` via Qdrant memory system
-- Lightweight analysis: Local models reduce API costs for repetitive patterns
+- Lightweight analysis: Local models reduce API costs for repetitive patterns

package/templates/skills/extended/backend/api-fuzzing-bug-bounty/SKILL.md

@@ -440,30 +440,35 @@ curl -X POST https://target.com/graphql \
 > **Adapted for [@techwavedev/agi-agent-kit](https://www.npmjs.com/package/@techwavedev/agi-agent-kit)**
 > Original source: [antigravity-awesome-skills](https://github.com/sickn33/antigravity-awesome-skills)
 
-### Qdrant Memory Integration
+### Hybrid Memory Integration (Qdrant + BM25)
 
 Before executing complex tasks with this skill:
 ```bash
 python3 execution/memory_manager.py auto --query "<task summary>"
 ```
+
+**Decision Tree:**
 - **Cache hit?** Use cached response directly — no need to re-process.
 - **Memory match?** Inject `context_chunks` into your reasoning.
 - **No match?** Proceed normally, then store results:
+
 ```bash
 python3 execution/memory_manager.py store \
   --content "Description of what was decided/solved" \
   --type decision \
-  --tags api-fuzzing-bug-bounty <relevant-tags>
+  --tags API Fuzzing for Bug Bounty <relevant-tags>
 ```
 
+> **Note:** Storing automatically updates both Vector (Qdrant) and Keyword (BM25) indices.
+
 ### Agent Team Collaboration
 
-- This skill can be invoked by the `orchestrator` agent via intelligent routing.
-- In **Agent Teams mode**, results are shared via Qdrant shared memory for cross-agent context.
-- In **Subagent mode**, this skill runs in isolation with its own memory namespace.
+- **Strategy**: This skill communicates via the shared memory system.
+- **Orchestration**: Invoked by `orchestrator` via intelligent routing.
+- **Context Sharing**: Always read previous agent outputs from memory before starting.
 
 ### Local LLM Support
 
 When available, use local Ollama models for embedding and lightweight inference:
 - Embeddings: `nomic-embed-text` via Qdrant memory system
-- Lightweight analysis: Local models reduce API costs for repetitive patterns
+- Lightweight analysis: Local models reduce API costs for repetitive patterns

package/templates/skills/extended/backend/async-python-patterns/SKILL.md

@@ -46,15 +46,18 @@ Refer to `resources/implementation-playbook.md` for detailed patterns and exampl
 > **Adapted for [@techwavedev/agi-agent-kit](https://www.npmjs.com/package/@techwavedev/agi-agent-kit)**
 > Original source: [antigravity-awesome-skills](https://github.com/sickn33/antigravity-awesome-skills)
 
-### Qdrant Memory Integration
+### Hybrid Memory Integration (Qdrant + BM25)
 
 Before executing complex tasks with this skill:
 ```bash
 python3 execution/memory_manager.py auto --query "<task summary>"
 ```
+
+**Decision Tree:**
 - **Cache hit?** Use cached response directly — no need to re-process.
 - **Memory match?** Inject `context_chunks` into your reasoning.
 - **No match?** Proceed normally, then store results:
+
 ```bash
 python3 execution/memory_manager.py store \
   --content "Description of what was decided/solved" \
@@ -62,14 +65,16 @@ python3 execution/memory_manager.py store \
   --tags async-python-patterns <relevant-tags>
 ```
 
+> **Note:** Storing automatically updates both Vector (Qdrant) and Keyword (BM25) indices.
+
 ### Agent Team Collaboration
 
-- This skill can be invoked by the `orchestrator` agent via intelligent routing.
-- In **Agent Teams mode**, results are shared via Qdrant shared memory for cross-agent context.
-- In **Subagent mode**, this skill runs in isolation with its own memory namespace.
+- **Strategy**: This skill communicates via the shared memory system.
+- **Orchestration**: Invoked by `orchestrator` via intelligent routing.
+- **Context Sharing**: Always read previous agent outputs from memory before starting.
 
 ### Local LLM Support
 
 When available, use local Ollama models for embedding and lightweight inference:
 - Embeddings: `nomic-embed-text` via Qdrant memory system
-- Lightweight analysis: Local models reduce API costs for repetitive patterns
+- Lightweight analysis: Local models reduce API costs for repetitive patterns

package/templates/skills/extended/backend/azure-ai-formrecognizer-java/SKILL.md

@@ -348,15 +348,18 @@ FORM_RECOGNIZER_KEY=<your-api-key>
 > **Adapted for [@techwavedev/agi-agent-kit](https://www.npmjs.com/package/@techwavedev/agi-agent-kit)**
 > Original source: [antigravity-awesome-skills](https://github.com/sickn33/antigravity-awesome-skills)
 
-### Qdrant Memory Integration
+### Hybrid Memory Integration (Qdrant + BM25)
 
 Before executing complex tasks with this skill:
 ```bash
 python3 execution/memory_manager.py auto --query "<task summary>"
 ```
+
+**Decision Tree:**
 - **Cache hit?** Use cached response directly — no need to re-process.
 - **Memory match?** Inject `context_chunks` into your reasoning.
 - **No match?** Proceed normally, then store results:
+
 ```bash
 python3 execution/memory_manager.py store \
   --content "Description of what was decided/solved" \
@@ -364,14 +367,16 @@ python3 execution/memory_manager.py store \
   --tags azure-ai-formrecognizer-java <relevant-tags>
 ```
 
+> **Note:** Storing automatically updates both Vector (Qdrant) and Keyword (BM25) indices.
+
 ### Agent Team Collaboration
 
-- This skill can be invoked by the `orchestrator` agent via intelligent routing.
-- In **Agent Teams mode**, results are shared via Qdrant shared memory for cross-agent context.
-- In **Subagent mode**, this skill runs in isolation with its own memory namespace.
+- **Strategy**: This skill communicates via the shared memory system.
+- **Orchestration**: Invoked by `orchestrator` via intelligent routing.
+- **Context Sharing**: Always read previous agent outputs from memory before starting.
 
 ### Local LLM Support
 
 When available, use local Ollama models for embedding and lightweight inference:
 - Embeddings: `nomic-embed-text` via Qdrant memory system
-- Lightweight analysis: Local models reduce API costs for repetitive patterns
+- Lightweight analysis: Local models reduce API costs for repetitive patterns

package/templates/skills/extended/backend/azure-communication-callingserver-java/SKILL.md

@@ -98,15 +98,18 @@ See the `azure-communication-callautomation-java` skill for:
 > **Adapted for [@techwavedev/agi-agent-kit](https://www.npmjs.com/package/@techwavedev/agi-agent-kit)**
 > Original source: [antigravity-awesome-skills](https://github.com/sickn33/antigravity-awesome-skills)
 
-### Qdrant Memory Integration
+### Hybrid Memory Integration (Qdrant + BM25)
 
 Before executing complex tasks with this skill:
 ```bash
 python3 execution/memory_manager.py auto --query "<task summary>"
 ```
+
+**Decision Tree:**
 - **Cache hit?** Use cached response directly — no need to re-process.
 - **Memory match?** Inject `context_chunks` into your reasoning.
 - **No match?** Proceed normally, then store results:
+
 ```bash
 python3 execution/memory_manager.py store \
   --content "Description of what was decided/solved" \
@@ -114,14 +117,16 @@ python3 execution/memory_manager.py store \
   --tags azure-communication-callingserver-java <relevant-tags>
 ```
 
+> **Note:** Storing automatically updates both Vector (Qdrant) and Keyword (BM25) indices.
+
 ### Agent Team Collaboration
 
-- This skill can be invoked by the `orchestrator` agent via intelligent routing.
-- In **Agent Teams mode**, results are shared via Qdrant shared memory for cross-agent context.
-- In **Subagent mode**, this skill runs in isolation with its own memory namespace.
+- **Strategy**: This skill communicates via the shared memory system.
+- **Orchestration**: Invoked by `orchestrator` via intelligent routing.
+- **Context Sharing**: Always read previous agent outputs from memory before starting.
 
 ### Local LLM Support
 
 When available, use local Ollama models for embedding and lightweight inference:
 - Embeddings: `nomic-embed-text` via Qdrant memory system
-- Lightweight analysis: Local models reduce API costs for repetitive patterns
+- Lightweight analysis: Local models reduce API costs for repetitive patterns

package/templates/skills/extended/backend/azure-cosmos-db-py/SKILL.md

@@ -246,15 +246,18 @@ async def test_get_project_by_id_returns_project(mock_cosmos_container):
 > **Adapted for [@techwavedev/agi-agent-kit](https://www.npmjs.com/package/@techwavedev/agi-agent-kit)**
 > Original source: [antigravity-awesome-skills](https://github.com/sickn33/antigravity-awesome-skills)
 
-### Qdrant Memory Integration
+### Hybrid Memory Integration (Qdrant + BM25)
 
 Before executing complex tasks with this skill:
 ```bash
 python3 execution/memory_manager.py auto --query "<task summary>"
 ```
+
+**Decision Tree:**
 - **Cache hit?** Use cached response directly — no need to re-process.
 - **Memory match?** Inject `context_chunks` into your reasoning.
 - **No match?** Proceed normally, then store results:
+
 ```bash
 python3 execution/memory_manager.py store \
   --content "Description of what was decided/solved" \
@@ -262,14 +265,16 @@ python3 execution/memory_manager.py store \
   --tags azure-cosmos-db-py <relevant-tags>
 ```
 
+> **Note:** Storing automatically updates both Vector (Qdrant) and Keyword (BM25) indices.
+
 ### Agent Team Collaboration
 
-- This skill can be invoked by the `orchestrator` agent via intelligent routing.
-- In **Agent Teams mode**, results are shared via Qdrant shared memory for cross-agent context.
-- In **Subagent mode**, this skill runs in isolation with its own memory namespace.
+- **Strategy**: This skill communicates via the shared memory system.
+- **Orchestration**: Invoked by `orchestrator` via intelligent routing.
+- **Context Sharing**: Always read previous agent outputs from memory before starting.
 
 ### Local LLM Support
 
 When available, use local Ollama models for embedding and lightweight inference:
 - Embeddings: `nomic-embed-text` via Qdrant memory system
-- Lightweight analysis: Local models reduce API costs for repetitive patterns
+- Lightweight analysis: Local models reduce API costs for repetitive patterns

package/templates/skills/extended/backend/azure-data-tables-java/SKILL.md

@@ -341,15 +341,18 @@ COSMOS_TABLE_ENDPOINT=https://<account>.table.cosmosdb.azure.com
 > **Adapted for [@techwavedev/agi-agent-kit](https://www.npmjs.com/package/@techwavedev/agi-agent-kit)**
 > Original source: [antigravity-awesome-skills](https://github.com/sickn33/antigravity-awesome-skills)
 
-### Qdrant Memory Integration
+### Hybrid Memory Integration (Qdrant + BM25)
 
 Before executing complex tasks with this skill:
 ```bash
 python3 execution/memory_manager.py auto --query "<task summary>"
 ```
+
+**Decision Tree:**
 - **Cache hit?** Use cached response directly — no need to re-process.
 - **Memory match?** Inject `context_chunks` into your reasoning.
 - **No match?** Proceed normally, then store results:
+
 ```bash
 python3 execution/memory_manager.py store \
   --content "Description of what was decided/solved" \
@@ -357,14 +360,16 @@ python3 execution/memory_manager.py store \
   --tags azure-data-tables-java <relevant-tags>
 ```
 
+> **Note:** Storing automatically updates both Vector (Qdrant) and Keyword (BM25) indices.
+
 ### Agent Team Collaboration
 
-- This skill can be invoked by the `orchestrator` agent via intelligent routing.
-- In **Agent Teams mode**, results are shared via Qdrant shared memory for cross-agent context.
-- In **Subagent mode**, this skill runs in isolation with its own memory namespace.
+- **Strategy**: This skill communicates via the shared memory system.
+- **Orchestration**: Invoked by `orchestrator` via intelligent routing.
+- **Context Sharing**: Always read previous agent outputs from memory before starting.
 
 ### Local LLM Support
 
 When available, use local Ollama models for embedding and lightweight inference:
 - Embeddings: `nomic-embed-text` via Qdrant memory system
-- Lightweight analysis: Local models reduce API costs for repetitive patterns
+- Lightweight analysis: Local models reduce API costs for repetitive patterns

package/templates/skills/extended/backend/azure-mgmt-apicenter-dotnet/SKILL.md

@@ -418,15 +418,18 @@ catch (RequestFailedException ex)
 > **Adapted for [@techwavedev/agi-agent-kit](https://www.npmjs.com/package/@techwavedev/agi-agent-kit)**
 > Original source: [antigravity-awesome-skills](https://github.com/sickn33/antigravity-awesome-skills)
 
-### Qdrant Memory Integration
+### Hybrid Memory Integration (Qdrant + BM25)
 
 Before executing complex tasks with this skill:
 ```bash
 python3 execution/memory_manager.py auto --query "<task summary>"
 ```
+
+**Decision Tree:**
 - **Cache hit?** Use cached response directly — no need to re-process.
 - **Memory match?** Inject `context_chunks` into your reasoning.
 - **No match?** Proceed normally, then store results:
+
 ```bash
 python3 execution/memory_manager.py store \
   --content "Description of what was decided/solved" \
@@ -434,14 +437,16 @@ python3 execution/memory_manager.py store \
   --tags azure-mgmt-apicenter-dotnet <relevant-tags>
 ```
 
+> **Note:** Storing automatically updates both Vector (Qdrant) and Keyword (BM25) indices.
+
 ### Agent Team Collaboration
 
-- This skill can be invoked by the `orchestrator` agent via intelligent routing.
-- In **Agent Teams mode**, results are shared via Qdrant shared memory for cross-agent context.
-- In **Subagent mode**, this skill runs in isolation with its own memory namespace.
+- **Strategy**: This skill communicates via the shared memory system.
+- **Orchestration**: Invoked by `orchestrator` via intelligent routing.
+- **Context Sharing**: Always read previous agent outputs from memory before starting.
 
 ### Local LLM Support
 
 When available, use local Ollama models for embedding and lightweight inference:
 - Embeddings: `nomic-embed-text` via Qdrant memory system
-- Lightweight analysis: Local models reduce API costs for repetitive patterns
+- Lightweight analysis: Local models reduce API costs for repetitive patterns