Using Agent Skills
Paste the following prompt into your AI chat to install this skill:
Please follow https://skillhub.cn/install/skillhub.md to install @user_3c6cb52e/using-agent-skills-lr-61zp.
About this skill
Problem
When tasks move across requirement clarification, implementation, and review feedback, engineers need the process guidance that matches the current phase. If all workflow rules are given to a model at once, the context becomes noisy and the answer may miss phase-critical behavioral constraints. Using Agent Skills addresses routing: first identify the development phase of the task, then select the corresponding Agent Skill.
How It Works
It behaves more like a phase navigation layer than a complete engineering framework. The core path is to read the current task goal, inputs, and constraints, map them to a development phase, apply the matching process instructions, and keep cross-phase core behaviors in scope. In practice, describe the task state and acceptance goals first, then ask which phase and skill should be used and what additional context is needed.
Boundaries
It depends on an existing set of phase-based Agent Skills. If a matching sub-skill is missing, it can only provide general engineering advice. It does not itself implement code, testing, or deployment actions, and it should not be treated as a fixed automated pipeline.
Use Cases
- After receiving a refactoring task, identify the implementation phase, select the matching workflow skill, and apply its process constraints before making changes.
- When requirement notes are incomplete, identify the clarification phase and choose the corresponding skill to structure the next engineering conversation.
- During incident review, identify diagnosis and remediation phases, then apply the matching skill to structure analysis, reproduction, and validation steps.
- Before delegating several engineering tasks to an agent, select the skill that fits the current phase and keep cross-phase core behaviors in scope.
Best For
- Agent engineers maintaining multi-phase engineering workflows need models to determine the task phase before choosing workflow guidance.
- Engineering leads implementing development standards need senior-engineer processes packaged into phase-based skills that agents can call.
- Engineers using Claude or similar agents for coding tasks need fewer skipped steps and less cross-phase process mixing.
- Platform developers building internal skill libraries need a meta-skill entry point for routing tasks to the right phase.
Related Skills
Local workflow memory with matching and SOP updates.
An OpenClaw live streaming executor that initializes TRTC streaming, starts a real-time dashboard, generates viewer URLs, and continuously reports live events.
Breaks down physical supply chains for super-trends to identify second- and third-layer bottlenecks, runs valuation and reverse checks, and maintains trackable reports.
A token-saving compression mode for Chinese LLMs with lite, full, ultra, and classical tiers, preserving code and technical terms while handling edge cases.