[02] META. Create new skills when existing ones don't cover the task. Analyze unique requirements, build framework (Frame → Research → Plan → Execute), integrate risks, and declare new skill. Use when facing novel problems that existing skills can't address.
68
Quality
51%
Does it follow best practices?
Impact
96%
1.62xAverage score across 3 eval scenarios
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./data/02-meta-skill-forge-150/SKILL.mdCore Principle: When existing skills don't fit, forge a new one. Analyze the unique task, build a proper framework, integrate specific risks, and create a reusable skill.
When you invoke this skill, you're asking AI to:
| Dimension | 100% Core | +50% Enhancement |
|---|---|---|
| Framework | Basic structure | + Task-specific adaptations |
| Criteria | Success metrics | + Quality checkpoints |
| Risks | Known challenges | + Mitigation strategies |
| Evolution | Current version | + Improvement path |
TRIGGER: Task is outside existing skill coverage:
DO NOT USE FOR:
FRAMEWORK FOUNDATION (100% Required)
├── Task Semantics: Core requirements analysis
├── Structure: Frame → Research → Plan → Execute
├── Criteria: Success and quality metrics
└── Validation: How to verify skill works
ADAPTATION LAYER (50% Enhancement)
├── Risk Integration: Specific challenge handling
├── Spirit Alignment: Task-appropriate approach
├── Resource Optimization: Efficient methods
└── Evolution Planning: Future improvements🔍 **GAP ANALYSIS**
**Task:** [What needs to be done]
**Existing Skills Checked:**
- [ ] goal-clarity-150: [Fits/Doesn't fit - why]
- [ ] research-deep-150: [Fits/Doesn't fit - why]
- [ ] impact-map-150: [Fits/Doesn't fit - why]
- [ ] action-plan-150: [Fits/Doesn't fit - why]
- [ ] chain-flow-150: [Fits/Doesn't fit - why]
- [ ] [Other skills...]: [Assessment]
**Gap Identified:** [What's missing]
**New Skill Justified:** ✅ Yes | ❌ No (use existing)📋 **UNIQUE REQUIREMENTS**
**What makes this task special:**
- [Unique aspect 1]
- [Unique aspect 2]
- [Unique aspect 3]
**Core challenges:**
- [Challenge 1]
- [Challenge 2]
**Success looks like:**
- [Success criterion 1]
- [Success criterion 2]
**Failure looks like:**
- [Failure mode 1]
- [Failure mode 2]🏗️ **SKILL FRAMEWORK**
**Skill Name:** [name-150]
**Purpose:** [One-line description]
**STRUCTURE:**
FRAME PHASE:
├── [Setup step 1]
├── [Setup step 2]
└── [Context establishment]
RESEARCH PHASE:
├── [Investigation step 1]
├── [Investigation step 2]
└── [Information synthesis]
PLAN PHASE:
├── [Strategy step 1]
├── [Strategy step 2]
└── [Preparation completion]
EXECUTE PHASE:
├── [Implementation step 1]
├── [Implementation step 2]
└── [Validation and completion]⚠️ **RISK HANDLING**
**Specific Risks:**
| Risk | Likelihood | Impact | Mitigation |
|------|------------|--------|------------|
| [Risk 1] | [H/M/L] | [H/M/L] | [Strategy] |
| [Risk 2] | [H/M/L] | [H/M/L] | [Strategy] |
**Built-in Safeguards:**
- [Safeguard 1]
- [Safeguard 2]
**Recovery Paths:**
- If [failure]: [recovery action]
- If [failure]: [recovery action]✅ **QUALITY CHECKPOINTS**
**Validation Methods:**
- [ ] [How to verify step 1 works]
- [ ] [How to verify step 2 works]
- [ ] [How to verify overall skill works]
**Success Metrics:**
- [Metric 1]: [How to measure]
- [Metric 2]: [How to measure]
**Quality Gates:**
- After Frame: [What must be true]
- After Research: [What must be true]
- After Plan: [What must be true]
- After Execute: [What must be true]⚡ **NEW SKILL DECLARED**
---
name: [skill-name-150]
description: [One-line description for skill discovery]
---
# [Skill Name] 150 Protocol
**Core Principle:** [Main idea in one sentence]
## What This Skill Does
[Bullet list of capabilities]
## The 150% Rule
[Table of 100% core + 50% enhancement]
## When to Use
[Trigger conditions]
## Execution Protocol
[Step-by-step process]
## Output Format
[Expected outputs]
## Operational Rules
[Numbered rules]
## Failure Modes & Recovery
[Table of failures and recoveries]
## Examples
[Good and bad examples]📝 **SKILL-FORGE 150 PROPOSAL**
**New Skill:** [name-150]
**Purpose:** [Description]
**Fills Gap:** [What was missing]
**Structure Summary:**
- Frame: [Brief]
- Research: [Brief]
- Plan: [Brief]
- Execute: [Brief]
**Approve skill creation?** (Yes / No / Modify)Proposal:
⚡ **SKILL-FORGE 150 PROPOSAL**
**Gap Identified:** [What existing skills don't cover]
**New Skill:**
├── Name: [skill-name-150]
├── Purpose: [One-line description]
└── Unique Value: [What it adds]
**Framework:**
├── Frame: [Summary]
├── Research: [Summary]
├── Plan: [Summary]
└── Execute: [Summary]
**Risks Addressed:**
- [Risk 1] → [Mitigation]
- [Risk 2] → [Mitigation]
**Approve creation?**Complete skill file ready to save.
---
name: [skill-name-150]
description: [One-line description that helps skill discovery. Include trigger words and use cases.]
---
# [Skill Name] 150 Protocol
**Core Principle:** [Main idea]
## What This Skill Does
When you invoke this skill, you're asking AI to:
- [Capability 1]
- [Capability 2]
- [Capability 3]
## The 150% Rule
| Dimension | 100% Core | +50% Enhancement |
|-----------|-----------|------------------|
| [Dim 1] | [Core] | [Enhancement] |
| [Dim 2] | [Core] | [Enhancement] |
## When to Use This Skill
- [Trigger 1]
- [Trigger 2]
- [Trigger 3]
## Execution Protocol
### Step 1: [NAME]
[Details]
### Step 2: [NAME]
[Details]
[Continue as needed]
## Output Format
[Template for skill output]
## Operational Rules
1. [Rule 1]
2. [Rule 2]
3. [Rule 3]
## Failure Modes & Recovery
| Failure | Detection | Recovery |
|---------|-----------|----------|
| [Failure 1] | [How to detect] | [How to fix] |
| [Failure 2] | [How to detect] | [How to fix] |
## Examples
### ❌ Without This Skill
[Bad example]
### ✅ With This Skill
[Good example]
---
**Remember:** [Key takeaway]| Failure | Detection | Recovery |
|---|---|---|
| Over-Engineering | Too complex for task | Simplify to essentials |
| Incomplete Coverage | Missing key aspects | Expand framework scope |
| Poor Adaptation | Doesn't fit task spirit | Re-align with requirements |
| No Validation | Can't verify it works | Add testing and feedback |
Task: Rename a variable
Gap: "No variable-rename-150 skill!"
Result: Created 20-step protocol for simple find-replace
Outcome: Over-engineering, wasted effort⚡ SKILL-FORGE 150 PROPOSAL
Gap Identified: No skill for zero-downtime data migrations
New Skill:
├── Name: live-migrate-150
├── Purpose: Execute data migrations without service interruption
└── Unique Value: Combines deployment, monitoring, and rollback
Framework:
├── Frame: Assess data volumes, set up shadow tables
├── Research: Analyze dependencies, identify sync points
├── Plan: Design phased migration with validation gates
└── Execute: Migrate incrementally with real-time verification
Risks Addressed:
- Data inconsistency → Dual-write with reconciliation
- Performance impact → Throttled batch processing
- Rollback needed → Point-in-time recovery points
Approve creation? Yes
[Full SKILL.md created and saved]Remember: Skill-forge is for genuine gaps, not for making simple things complex. A well-forged skill becomes a reusable asset. A poorly-forged skill becomes technical debt. Forge wisely.
2bbaa03
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.