zuhair-01/claude-skills-and-systems/prompts-chat
Claude Code and agent skill by zuhair-01/claude-skills-and-systems
- community
Checking that the SKILL.md file is still in the repository.
2,169 categorized prompts from github.com/f/prompts.chat, searched off-context. Use to find a proven persona/task/system prompt before writing one ("act as X", "prompt for", system prompts).
Install
npx vskill@latest install zuhair-01/claude-skills-and-systems/prompts-chatSKILL.md
The same SKILL.md was published earlier as zuhair-01/claude-skills-and-systems/prompts-chat.
prompts.chat library (off-context, search first)
Data is NOT loaded into context. Query it:
Per-category title lists: categories/<name>.md. Raw data: prompts.jsonl (id, act, category, tags, type, devs, prompt).
Categories (18): code-dev, image-gen, general-utility, creative-media, writing-content, data-ai, marketing-business, education-tutoring, finance-trading, roleplay-characters, design-ui, devops-security, career-hr, advice-life, health-wellness, legal-gov, language-translation, jailbreak-unsafe (quarantined).
How to use smartly
- Search 2-3 keyword variants;
--showthe top 1-3. Read them for structure (role, constraints, output format, first-turn behaviour), not just wording. - Adapt, don't paste: tailor to the real task/audience/language, drop filler ("I want you to act as..." boilerplate), add explicit output format and success criteria.
- Combine with
prompt-master(sharpen the final prompt),agentic-system-prompt-patterns(agent system prompts),marketing-psychology(copy personas),ugc-*/image skills (useimage-genand STRUCTURED entries as templates). jailbreak-unsafeis quarantined: excluded from search, never used to bypass safeguards; reference only for defensive/red-team study.- Cite the source (prompts.chat, CC0) when shipping a derived prompt to a client.
Refresh
cd ~/.claude/prompts-library-src/pc && git pull then python ~/.claude/skills/prompts-chat/build.py.
Auto-trigger
overseer/prompts_chat_gate.py fires only when (1) intent = persona/prompt ask or write/draft/make + deliverable, (2) not ops/meta talk (deploy, git, hooks, bugs...), (3) whole-word title match (2+ words or 1 rare word), max 3 hits, deduped per session. English only; tune EXPLICIT/CREATE/OPS/STOP lists at the top of the file.
Security
- Security scanPassed, 95 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingskills/prompts-chat/SKILL.md
Popularity
- Repository stars
- 1
- 7-day trend score
- 8
Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/zuhair-01/claude-skills-and-systems/prompts-chat)