catlog22/Claude-Code-Workflow
1
0
Fork 0
mirror of https://github.com/catlog22/Claude-Code-Workflow.git synced 2026-02-12 02:37:45 +08:00
Code Issues Packages Projects Releases Wiki Activity

feat: implement dynamic SKILL discovery with intelligent matching

Browse Source 
Transform load-skill-memory from manual specification to automatic discovery:

**Core Change**:
- From: User specifies SKILL name manually
- To: System automatically discovers and matches SKILL based on task context

**New Capabilities**:

1. **Three-Step Execution**:
   - Step 1: Discover all available SKILLs (.claude/skills/)
   - Step 2: Match most relevant SKILL using scoring algorithm
   - Step 3: Activate matched SKILL via Skill() tool

2. **Intelligent Matching Algorithm**:
   - **Path-Based** (Highest Priority): Direct path match from file paths
   - **Keyword Matching** (Secondary): Score by keyword overlap
   - **Action Matching** (Tertiary): Detect action verbs (分析/修改/学习)

3. **Updated Parameters**:
   - From: `<skill_name> [--level] [task description]`
   - To: `"task description or file path"`
   - More intuitive user experience

4. **New Examples**:
   ```bash
   /memory:load-skill-memory "分析热模型builder pattern实现"
   /memory:load-skill-memory "D:\dongdiankaifa9\hydro_generator_module\builders\base.py"
   /memory:load-skill-memory "修改workflow文档生成逻辑"
   ```

**Matching Examples**:

Task: "分析热模型builder pattern实现"
- hydro_generator_module: 4 points (thermal+builder+analyzing) ✅
- Claude_dms3: 1 point (analyzing only)

Task: "D:\dongdiankaifa9\hydro_generator_module\builders\base.py"
- Path match: hydro_generator_module ✅ (exact path)

**Benefits**:
- No manual SKILL name required
- Automatic best match selection
- Path-based intelligent routing
- Keyword scoring for relevance
- Action verb detection for context

**User Experience**:
Before: "/memory:load-skill-memory hydro_generator_module '分析热模型'"
After: "/memory:load-skill-memory '分析热模型实现'"

System automatically discovers and activates hydro_generator_module SKILL.

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>
This commit is contained in:
catlog22
2025-11-03 21:31:06 +08:00
parent 9e0a70150a
commit 25a358b729
1 changed files with 172 additions and 78 deletions
Download Patch File Download Diff File Expand all files Collapse all files

248
.claude/commands/memory/load-skill-memory.md
View File

@@ -1,78 +1,126 @@
--- ---
name: load-skill-memory name: load-skill-memory
description: Load SKILL package documentation with progressive levels based on task requirements description: Automatically discover and activate SKILL packages based on task context with intelligent matching
argument-hint: "<skill_name> [--level 0|1|2|3] [task description]" argument-hint: "\"task description or file path\""
allowed-tools: Bash(*), Read(*), Glob(*), Skill(*) allowed-tools: Bash(*), Read(*), Glob(*), Skill(*)
examples: examples:
- Skill(command: "hydro_generator_module") - /memory:load-skill-memory "分析热模型builder pattern实现"
- Skill(command: "Claude_dms3") - /memory:load-skill-memory "修改workflow文档生成逻辑"
- Skill(command: "multiphysics_network") - /memory:load-skill-memory "D:\dongdiankaifa9\hydro_generator_module\builders\base.py"
--- ---
# Memory Load SKILL Command (/memory:load-skill-memory) # Memory Load SKILL Command (/memory:load-skill-memory)
## 1. Overview ## 1. Overview
The `memory:load-skill-memory` command **activates SKILL packages** generated by `/memory:skill-memory`. It validates SKILL existence and uses the `Skill()` tool to load comprehensive documentation context. The `memory:load-skill-memory` command **automatically discovers and activates** the most relevant SKILL package based on task description or file path. It dynamically matches user intent against available SKILL descriptions and activates the best match.
**Core Philosophy**: **Core Philosophy**:
- **Two-Step Process**: Validate SKILL existence → Activate via Skill tool - **Dynamic SKILL Discovery**: Automatically finds relevant SKILL without manual specification
- **Automatic Context Loading**: SKILL system loads appropriate documentation automatically - **Intelligent Matching**: Matches task keywords against SKILL descriptions
- **No Manual Reading**: Use Skill tool instead of manual Read operations - **Path-Based Detection**: Recognizes project paths and activates corresponding SKILL
- **Task-Driven Activation**: System automatically determines required context based on SKILL description triggers - **Automatic Activation**: Uses Skill() tool to load comprehensive context
## 2. Parameters ## 2. Parameters
- `<skill_name>` (Required): Name of SKILL package to activate - `"task description or file path"` (Required): Task context or file path for SKILL matching
- Example: `hydro_generator_module` - **Task description**: "分析热模型实现", "修改workflow逻辑", "学习参数系统"
- Example: `Claude_dms3` - **File path**: "D:\dongdiankaifa9\hydro_generator_module\builders\base.py"
- Example: `multiphysics_network` - **Domain keywords**: "thermal modeling", "workflow management", "multi-physics"
## 3. Two-Step Execution Flow ## 3. Three-Step Execution Flow
### Step 1: Validate SKILL Existence ### Step 1: Discover Available SKILLs
**Check SKILL.md File**: **List All SKILL Packages**:
```bash
bash(test -f .claude/skills/{skill_name}/SKILL.md && echo "exists" || echo "not_found")
```
**Validation Logic**:
- If `exists`: Continue to Step 2
- If `not_found`: Return error message with available SKILLs:
```bash ```bash
bash(ls -1 .claude/skills/ 2>/dev/null || echo "No SKILLs available") bash(ls -1 .claude/skills/ 2>/dev/null || echo "No SKILLs available")
``` ```
**Error Message Format**: **Read Each SKILL.md for Matching**:
```bash
# For each SKILL directory found
for skill in $(ls -1 .claude/skills/); do
Read(.claude/skills/${skill}/SKILL.md)
done
``` ```
❌ SKILL '{skill_name}' not found.
**Extract Matching Information**:
- SKILL name
- Description (with trigger keywords)
- Location path (from description)
- Domain keywords
### Step 2: Match Most Relevant SKILL
**Matching Algorithm**:
1. **Path-Based Matching** (Highest Priority):
- Extract path from user input if provided
- Compare against SKILL location paths in descriptions
- Exact match: `D:\dongdiankaifa9\hydro_generator_module` → `hydro_generator_module` SKILL
2. **Keyword Matching** (Secondary Priority):
- Extract keywords from task description
- Match against SKILL description keywords
- Score each SKILL by keyword overlap count
3. **Action Matching** (Tertiary Priority):
- Detect action verbs: "分析", "修改", "学习", "实现"
- Match against SKILL description triggers
- Prefer SKILLs with matching action patterns
**Scoring Example**:
```javascript
Task: "分析热模型builder pattern实现"
hydro_generator_module SKILL:
- Path match: No
- Keywords: "热模型"(1), "builder"(1), "实现"(1) = 3 points
- Action match: "analyzing"(1) = 1 point
- Total: 4 points ✅ Winner
Claude_dms3 SKILL:
- Path match: No
- Keywords: "workflow"(0) = 0 points
- Action match: "analyzing"(1) = 1 point
- Total: 1 point
```
**No Match Handling**:
```
⚠️ No matching SKILL found for: "{task_description}"
Available SKILLs: Available SKILLs:
- hydro_generator_module (D:\dongdiankaifa9\hydro_generator_module) - hydro_generator_module - Hydro-generator thermal modeling
- Claude_dms3 (D:\Claude_dms3) - Claude_dms3 - Workflow orchestration system
Use: Skill(command: "skill_name") Generate SKILL for your project: /memory:skill-memory [path]
Generate new SKILL: /memory:skill-memory [path]
``` ```
### Step 2: Activate SKILL Package ### Step 3: Activate Matched SKILL
**Use Skill Tool to Activate**: **Activate Best Match**:
```javascript ```javascript
Skill(command: "{skill_name}") Skill(command: "{matched_skill_name}")
``` ```
**What Happens**: **What Happens**:
1. Claude's SKILL system reads `.claude/skills/{skill_name}/SKILL.md` 1. System reads `.claude/skills/{matched_skill_name}/SKILL.md`
2. Automatically loads appropriate documentation based on: 2. Automatically loads appropriate documentation based on:
- SKILL description triggers (analyzing, modifying, learning) - SKILL description triggers (analyzing, modifying, learning)
- Current conversation context - Current conversation context
- Memory gaps detection - Memory gaps detection
3. Progressive loading levels (0-3) handled automatically by system 3. Progressive loading levels (0-3) handled automatically
4. Context loaded directly into conversation memory 4. Context loaded directly into conversation memory
**No Manual Reading Required**: The Skill tool handles all documentation loading automatically based on SKILL.md configuration. **Confirmation Output**:
```
✅ Matched and activated SKILL: {matched_skill_name}
🎯 Match reason: {path/keyword/action match}
📦 Location: {project_path}
💡 Context loaded for: {domain_description}
```
## 4. Output Format ## 4. Output Format
@@ -95,63 +143,109 @@ Skill(command: "{skill_name}")
## 5. Usage Examples ## 5. Usage Examples
### Example 1: Activate Hydro Generator Module SKILL ### Example 1: Task-Based Discovery (Keyword Matching)
**User Request**: "分析热模型builder pattern实现" **User Command**:
```bash
/memory:load-skill-memory "分析热模型builder pattern实现"
```
**Command Execution**: **Execution Flow**:
```javascript ```javascript
// Step 1: Validate // Step 1: Discover available SKILLs
bash(test -f .claude/skills/hydro_generator_module/SKILL.md && echo "exists") bash(ls -1 .claude/skills/)
// Output: exists // Output: hydro_generator_module, Claude_dms3
// Step 2: Activate // Read each SKILL.md
Read(.claude/skills/hydro_generator_module/SKILL.md)
Read(.claude/skills/Claude_dms3/SKILL.md)
// Step 2: Match keywords
Keywords extracted: ["热模型", "builder", "pattern", "实现", "分析"]
Matching scores:
- hydro_generator_module: 4 points (thermal modeling, builder, analyzing)
- Claude_dms3: 1 point (analyzing only)
Best match: hydro_generator_module
// Step 3: Activate matched SKILL
Skill(command: "hydro_generator_module") Skill(command: "hydro_generator_module")
``` ```
**SKILL Activation**: **Output**:
- System reads `.claude/skills/hydro_generator_module/SKILL.md` ```
- Description triggers match: "analyzing" + "hydro-generator thermal modeling" ✅ Matched and activated SKILL: hydro_generator_module
- Automatically loads appropriate documentation (Level 1-2) 🎯 Match reason: Keywords ["thermal", "builder"] + Action ["analyzing"]
- Context available for subsequent analysis 📦 Location: D:\dongdiankaifa9\hydro_generator_module
💡 Context loaded for: hydro-generator thermal modeling
```
### Example 2: Activate Workflow System SKILL ### Example 2: Path-Based Discovery (Direct Path Match)
**User Request**: "修改文档生成workflow调度逻辑" **User Command**:
```bash
/memory:load-skill-memory "D:\dongdiankaifa9\hydro_generator_module\builders\base.py"
```
**Command Execution**: **Execution Flow**:
```javascript ```javascript
// Step 1: Validate // Step 1: Discover SKILLs
bash(test -f .claude/skills/Claude_dms3/SKILL.md && echo "exists") bash(ls -1 .claude/skills/)
// Output: exists
// Step 2: Activate // Step 2: Match path
Path extracted: "D:\dongdiankaifa9\hydro_generator_module"
Matching:
- hydro_generator_module location: "D:\dongdiankaifa9\hydro_generator_module" ✅ Exact match
- Claude_dms3 location: "D:\Claude_dms3" ❌ No match
Best match: hydro_generator_module (path match - highest priority)
// Step 3: Activate
Skill(command: "hydro_generator_module")
```
**Output**:
```
✅ Matched and activated SKILL: hydro_generator_module
🎯 Match reason: Path match (D:\dongdiankaifa9\hydro_generator_module)
📦 Location: D:\dongdiankaifa9\hydro_generator_module
💡 Context loaded for: hydro-generator thermal modeling
```
### Example 3: Domain Keyword Discovery
**User Command**:
```bash
/memory:load-skill-memory "修改workflow文档生成调度逻辑"
```
**Execution Flow**:
```javascript
// Step 1: Discover SKILLs
bash(ls -1 .claude/skills/)
// Step 2: Match keywords
Keywords: ["workflow", "文档生成", "调度", "修改"]
Matching scores:
- Claude_dms3: 3 points (workflow, docs generation, modifying)
- hydro_generator_module: 1 point (modifying only)
Best match: Claude_dms3
// Step 3: Activate
Skill(command: "Claude_dms3") Skill(command: "Claude_dms3")
``` ```
**SKILL Activation**: **Output**:
- Description triggers match: "modifying" + "workflow management" ```
- System loads comprehensive documentation (Level 2-3) ✅ Matched and activated SKILL: Claude_dms3
- Ready for workflow modification tasks 🎯 Match reason: Keywords ["workflow", "docs"] + Action ["modifying"]
📦 Location: D:\Claude_dms3
### Example 3: Activate Multi-Physics Network SKILL 💡 Context loaded for: workflow orchestration and documentation
**User Request**: "学习电磁域组件设计模式"
**Command Execution**:
```javascript
// Step 1: Validate
bash(test -f .claude/skills/multiphysics_network/SKILL.md && echo "exists")
// Output: exists
// Step 2: Activate
Skill(command: "multiphysics_network")
``` ```
**SKILL Activation**:
- Description triggers match: "learning" + "multi-physics"
- System provides overview and core module docs (Level 0-1)
- Context available for exploration
## 6. SKILL Trigger Mechanism ## 6. SKILL Trigger Mechanism