Prompt Master CLI
Paste the following prompt into your AI chat to install this skill:
Follow https://skillhub.cn/install/skillhub.md to install @user_5d00e27d/majia-system-prompt-cli.
About this skill
Problem
Engineers often know the desired output but struggle to express it as an executable LLM instruction: prompts are vague, missing constraints, or misclassified by task type. majia-system-prompt-cli is a local CLI workflow that turns fuzzy intent into structured prompt tasks rather than replacing the model.
How It Works
It first applies safety and intent checks, then routes work by task shape:
- M1 Research: ask for goals, targets, and constraints on first-time or ambiguous domains
- M2/M3: apply templates or expand creative tasks
- M4: analyze provided materials or data
- M5: decompose multi-step or cross-domain workflows
During execution it loads rules from rules/ for safety, research, output quality, and preference evolution, then runs a confidence self-check before output. Lightweight caching stores stable preferences to reduce repetitive clarification.
Boundaries
It is useful for prompt structuring, task decomposition, and output constraints, but it is not a compliance authority. Legal, medical, safety, and copyright cases still require human review, and output quality depends on the configured provider and model in config.yaml.
Use Cases
- Expand “write a marketing plan” in a terminal into a prompt with goals, audience, channels, and deliverables.
- Switch between DeepSeek and Claude via `config.yaml` to compare outputs for the same local analysis task.
- Use M4 analysis to cluster user feedback, then produce actionable improvement recommendations.
- Trigger M1 research on a new domain to confirm goals, constraints, and acceptance criteria before generation.
Best For
- AI application developers who need to turn vague requirements into executable prompts with fewer rewrites.
- Prompt engineers debugging multi-model CLI outputs who compare model differences and constraint adherence.
- Independent researchers who want to persist personal preferences and reduce repeated goal formatting.
- Automation engineers building CLI workflows who need generation, analysis, and planning routed into steps.
Related Skills
A guide for writing Seedance 2.0 video prompts, covering @ references, camera moves, time segments, audio direction, and common pitfalls.
Collects daily conversations, task logs, and errors, then runs a nine-module review to surface blind spots, generate actions, and update memory.
Local long-term memory for AI agents that appends key facts and daily logs, supports cross-session context and search recall, and auto-archives old diaries to gzip files.
A hybrid-evolution agent for OpenClaw developers that offers tiered-confirmation and rollback-safe capability optimization.