出海数字营销宝典

AI Agent Prompt Best Practices

Plug-and-play description skeletons and hierarchical router prompts to boost multi-agent system hit rates above 95%.

Agent Skill Description Template
Agent Recall & Reranking Router Prompt
Structured Description

Agent Skill Description Template

Author: YunxiEst. Read Time: 2 min

Description

Standardize individual Skill triggers, scopes, and negative samples to improve physical accuracy and target selection.

Key Benefits

  • Define legal trigger conditions
  • Add negative testcases to prevent confusion
  • Define input and output expectations
system_prompt.md
Use this Skill when the user needs to perform [Specific Task].

Applicable Scenarios:
- [Scenario 1]
- [Scenario 2]
- [Scenario 3]

Non-applicable Scenarios:
- [Unsuitable Task 1]
- [Unsuitable Task 2]
- [Unsuitable Task 3]

Input Information Typically Includes:
- [Information provided by user 1]
- [Information provided by user 2]
- [Information provided by user 3]

Output Requirements:
- [Output Format 1]
- [Output Format 2]
- [Output Style 3]

Recommended Reading List

Understanding recall, orchestration, and hierarchical decision-making in multi-agent systems.

《Design Patterns for LLM Agents》

Analyzing the core pathway and data structures from single agent systems to multi-agent collaboration and recall trees.

《Systematic Prompt Engineering Handbook》

Exploring system prompts, role control, and negative constraints refinement.

Explore / AI Agent Best Practices

AI Agent & Prompt Engineering Best Practices

A collection of cutting-edge guides, operational manuals, and optimization formulas for LLM Agents and multi-skill routing accuracy.