Guides
PCL-Lite
The World's First Programming Language for AI Persona Management
Make AI behavior programmable, portable, and predictable.
This page is the index of usage guides: how to use each bootstrap and where to generate or paste output (copy-paste or same chat).
PCL-Lite always gives you text in the chat: prompts, scripts, specs, structure. What you do next depends on your AI:
- If your AI doesn't generate image / video / audio / slides: copy the output from the chat and paste it into a dedicated tool (DALL·E, Runway, ElevenLabs, Google Slides, etc.). Each guide lists where to paste with direct links.
-
If your AI is multimodal (e.g. ChatGPT with DALL·E, Gemini with image/video, or any chat that can generate image/video/audio/slides in the same session): you can use the PCL-Lite output in the same chat. Ask for a prompt or script with
/persona, then paste that prompt or script into the image/video/audio/slides feature of the same AI.
See Text-only vs multimodal for details.
Workflow (text-only AI): Chat (PCL-Lite) → get text → copy → paste into external tool → get asset.
Workflow (multimodal AI): Chat (PCL-Lite) → get prompt/script/specs → use it in the same chat or copy-paste to an external tool — your choice.
| Guide | Bootstrap | Use case | Where to paste output |
|---|---|---|---|
| Standard | bootstrap_lite_en.md / FR | Architecture, dev, DevOps, teams | No paste step (text only) |
| Image | bootstrap-image/ | Visual design, illustrations, brand, a11y | DALL·E, Midjourney, Ideogram, etc. — see guide |
| Video | bootstrap-video/ | Scripts, storyboards, captions, delivery | Runway, Synthesia, Descript, CapCut, etc. — see guide |
| Audio | bootstrap-audio/ | Podcast, voiceover, sound design, transcripts | ElevenLabs, Descript, Murf, etc. — see guide |
| Slides | bootstrap-slides/ | Presentations, pitch decks, speaker notes | Google Slides, Gamma, Canva, etc. — see guide |
- Choose the bootstrap that matches your task (Standard, Image, Video, Audio, or Slides).
- Open the corresponding guide and the bootstrap file (same language: EN or FR) from the repo.
- Copy the entire bootstrap into your AI chat (first message).
- Wait for confirmation (e.g. "PCL-Lite … Ready").
- Use
/personacommands and ask for what you need (prompt, script, deck structure, etc.). - If your AI can generate (image/video/audio/slides): use that output in the same chat. If not: copy the output and paste it into the tool linked in the guide.
Chats: ChatGPT, Claude, Gemini, DeepSeek, Cursor — paste the bootstrap, then use /persona commands. The chat gives you text (prompts, scripts, specs). If your chat is multimodal, use that text in the same chat; otherwise each guide lists where to generate (with links) so you can copy-paste into an external tool.
- Examples to Try — Copy-paste examples to verify each bootstrap works.
- One bootstrap per conversation — Why we recommend one bootstrap per chat.
- Text-only vs multimodal — Same chat or external tool.
- Home