yetone
Featured
kill-ai-slop
Kill AI Slop is a multilingual field guide and agent skill that catalogues 33 common AI-generated design tells with interactive before-after demos, and provides a tool to scan and fix web projects.
AI repository intelligence
FR-AI / ANALYSISWhy it is worth attention
It turns generic chatbot prompt writing into a disciplined, business-vertical process with explicit boundaries, output contracts, and self-tests, which directly supports real deployment needs.
Who it is for
- Developers building domain-specific AI agents
- Prompt and system engineers
- Teams using npx skills or monorepo skill workflows
- People shipping support, sales, or internal-ops agents
Use cases
- Generate a system prompt for a customer-support agent with refusal rules and tool boundaries
- Create sales or internal-ops agent prompts with configured domain, tools, and strictness
- Add output contracts and self-tests before regression-testing an agent
- Produce bilingual prompt variants using the lang option
Strengths
- Concrete deliverables: system prompt, refusal table, tool boundary card, output contract, and 10 tests
- Configurable via domain, tools, strictness, length, and language
- Explicitly avoids generic 'helpful assistant' mush
- Simple install via npx skills add or local monorepo
Considerations
- README provides no benchmarks or extensive production examples
- Adoption assumes the skills CLI or monorepo setup
- Scope is prompt generation, not runtime behavior or legal advice
README quick start
Agent System Prompt
Vertical business system prompt: role, tool boundaries, refusals, output contract, plus 10 self-tests.
中文简介： 垂直业务 System Prompt（角色/工具边界/拒答/输出契约 + 自测 10 例）。 — 详见 README.zh.md.
What this skill does
- Writes a paste-ready system prompt for one vertical agent
- Adds refusal table, tool boundary card, and output contract
- Includes 10 self-test cases with pass/fail signals
- Avoids generic “helpful assistant” mush
Install
npx skills add / --skill agent-system-prompt
Local monorepo: skills/agent-system-prompt/.
Options
| Option | Values | Default |
|---|---|---|
lang | en · zh · bilingual | en |
domain | support · sales · internal-ops · devtools · custom | custom |
tools | none · read · read-write | read |
strictness | friendly · neutral · strict | neutral |
length | compact · standard · full | standard |
When to use
| Use | Don't |
|---|---|
| Ship a domain agent with policies | Generic chatbot with no vertical |
| Need refusals + tool limits written down | Support-only macro pack alone |
| Self-test before regression pack | Legal advice substitute |
Deliverable
System prompt + refusal table + output contract + tool card + 10 tests.
See SKILL.md.
Example
../../output/samples/agent-system-prompt-example-01.md
License
MIT — see LICENSE.
Related repositories
Similar projects matched by category, topics, and programming language.
gnipbao
Featured
story-to-handdrawn-video
A Remotion-based tool that converts Chinese story text or ordered hand-drawn images into vertical hand-drawn diary-comic animation with handwritten captions, left-to-right reveals, optional page-curl transitions, and silent H.264 output for post-production dubbing.
MoonshotAI
Featured
Kimi-K3
Kimi K3 is an open-weight, 2.8T-parameter native multimodal agentic model with a 1M-token context window, designed for frontier coding, knowledge work, and reasoning tasks.