From tac
Guides prompt level selection for tasks using seven levels framework, decision tree, complexity indicators, and examples to match investment to needs.
npx claudepluginhub melodic-software/claude-code-plugins --plugin tacThis skill is limited to using the following tools:
Guide selection of the appropriate prompt level for a task using the seven levels framework.
Displays seven levels of agentic prompts reference with table, decision guide, 80/20 rule, and related commands to select prompt complexity.
Crafts or updates LLM prompts from first principles by discovering goals, constraints, and context through targeted questions. Use for new prompts, updates, or reviews.
Optimizes prompts for production AI features with analysis, 6-step framework, failure detection, and research-backed techniques. Use for prompt review, system prompts, or improvement suggestions.
Share bugs, ideas, or general feedback.
Guide selection of the appropriate prompt level for a task using the seven levels framework.
Match task complexity to the right prompt level. Start simple, add complexity only when needed.
| Level | Name | Use When |
|---|---|---|
| 1 | High-Level | Simple repeatable task |
| 2 | Workflow | Sequential steps needed |
| 3 | Control Flow | Conditionals or loops |
| 4 | Delegation | Multiple agents needed |
| 5 | Higher-Order | Processing other prompts |
| 6 | Template Meta | Generating prompts |
| 7 | Self-Improving | Knowledge accumulation |
Is this a simple, repeatable task?
├── Yes -> Level 1 (High-Level Prompt)
└── No
|
Does it need sequential steps?
├── Yes -> Does it need conditionals/loops?
| ├── Yes -> Does it delegate to agents?
| | ├── Yes -> Level 4 (Delegation)
| | └── No -> Level 3 (Control Flow)
| └── No -> Level 2 (Workflow)
└── No
|
Does it process other prompts?
├── Yes -> Level 5 (Higher-Order)
└── No
|
Does it generate prompts?
├── Yes -> Level 6 (Template Meta)
└── No
|
Does it need to learn over time?
├── Yes -> Level 7 (Self-Improving)
└── No -> Reassess requirements
"Levels 3-4 cover 80% of practical use cases."
| Level Range | Coverage | Complexity |
|---|---|---|
| 1-2 | 40% | Low |
| 3-4 | 80% | Medium (Sweet Spot) |
| 5-7 | 20% | High |
Don't over-engineer. Most tasks fit in Levels 3-4.
Ask:
| Indicator | Points To |
|---|---|
| One-time or rare | Level 1 |
| Sequential steps | Level 2+ |
| "If X then Y" | Level 3+ |
| "Run N agents" | Level 4 |
| "Process this spec file" | Level 5 |
| "Create prompts for..." | Level 6 |
| "Learn and improve" | Level 7 |
Analysis: Simple, repeatable, no variables Level: 1 (High-Level Prompt)
Analysis: Sequential steps, needs input, produces output Level: 2 (Workflow Prompt)
Analysis: Loop required, conditional checking Level: 3 (Control Flow)
Analysis: Multiple agents, aggregation needed Level: 4 (Delegation Prompt)
Analysis: Accepts another prompt/spec as input Level: 5 (Higher-Order)
Analysis: Generates prompts in specific format Level: 6 (Template Meta Prompt)
Analysis: Accumulates expertise over time Level: 7 (Self-Improving Prompt)
When recommending a level:
## Level Selection
**Task:** [description]
**Recommended Level:** [1-7] ([name])
**Rationale:**
- [reason 1]
- [reason 2]
**Key Sections Needed:**
- [section 1]
- [section 2]
**Alternative Consideration:**
Level [N] if [condition]
| Red Flag | Issue | Solution |
|---|---|---|
| Jumping to Level 6-7 | Over-engineering | Start at Level 2-3 |
| Level 1 for complex task | Under-engineering | Add Workflow section |
| Level 4 for single-agent | Unnecessary delegation | Use Level 2-3 |
| No clear level fit | Vague requirements | Clarify task scope |
"Three times marks a pattern. Copy whatever you're doing and write it as a high level prompt, then move up the levels from there."
Date: 2025-12-26 Model: claude-opus-4-5-20251101