glunty AI tool cheat sheets / ChatGPT
ChatGPT hacks for power users
ChatGPT is OpenAI's assistant, and its real edge is the ecosystem around it: image generation, web browsing, and file tools. Power users lean on Custom Instructions, Projects, custom GPTs, and Deep Research rather than one endless chat. This guide covers the features and hacks that separate casual use from a real workflow.
Last verified: 2026-07-06. Free tiers and features change, so check the official site.
At a glance
- Maker
- OpenAI.
- Best for
- An all-in-one assistant with image generation, web browsing, and file tools.
- Power features
- Custom Instructions, Projects, custom GPTs, and Deep Research.
- Its edge
- The surrounding ecosystem, not just the raw model.
Features and hacks
Custom Instructions Persistent behavior across chats.
Example: Split "how to respond" from "who I am".
Projects Scoped memory and files.
Example: One project per client or domain.
GPTs Specialized workflows.
Example: Only build one when instructions repeat daily.
Deep Research Long, sourced reports.
Example: Start narrow and add sources iteratively.
Canvas Co-edit a doc or code.
Example: "Edit the doc, do not rewrite from scratch."
Memory Cross-chat preferences.
Example: Toggle it off for sensitive work.
Voice mode Hands-free use.
Example: Good for brainstorming, not precise code.
Web browse Current events and live data.
Example: Verify citations manually.
No rows match that search. Clear the box to see all 8.
Tips
Projects over endless chats
Use a Project instead of one endless chat so you can reset context without losing your files and instructions.
Rules, not essays
Keep Custom Instructions to concrete rules. A short list of do and do-not beats a long paragraph about yourself.
Ask for structured output
Request a table or JSON when the result feeds a downstream tool, so you can paste it straight in without cleanup.
Format first for research
For Deep Research, give the final format up front, like a memo, a slide outline, or an FAQ, so the report lands usable.
Play to the ecosystem
Do not duplicate another model's strength. Reach for ChatGPT when you want image generation, browsing, or Office file tools.
GPTs only when it repeats
Build a custom GPT only when the same instructions come up daily. Otherwise a Project or a saved prompt is simpler.
Operator recipe
You will make: A reusable ChatGPT Project plus custom instructions that drafts on-brand client emails from one short prompt.
- Set custom instructions in two halves: who you are (context) and how ChatGPT should respond (tone, format, length).
- Create one Project per client or domain so files and memory stay scoped and do not leak between clients.
- Write Project instructions with a clear role and an exact output format; these override your global custom instructions.
- Ground the Project: upload the brand guide, a past email, and any brief up to the free file cap (verify current limit).
- Draft cheap: generate a first email, then change one lever per turn (tone, length, or structure) instead of rewriting everything.
- Stuck on prompt wording? Route to /ai-prompts/prompting-basics or paste the prompt into the /improve-prompt tool for a tighter version.
- Save the winning email back to the Project as a source so it becomes a reusable on-brand reference.
Pass check: You can open the Project, type a one-line request, and get an on-brand email that needs only light edits.
Failure modes
| Symptom | Cause | Fix |
|---|---|---|
| ChatGPT states facts or prices you never gave it | The model fills gaps with plausible guesses | Ground it with pasted sources and add "never invent, say verify live" to Project instructions. |
| Drafts sound generic and off-brand | No voice reference or role was given | Upload the brand guide and a past deliverable, then set the role in Project instructions. |
| It forgets context from an earlier chat | Context is not shared between separate chats | Work inside one Project and save good responses back as sources. |
| One client's data appears in another client's chat | No isolation between clients | Use a separate Project per client with project-only memory (verify current features). |
| Long research doc reads vague and rambling | Unclear ask, plus too many changes at once | State the output format first, then change one lever per iteration. |
Workflow ladder
Set Custom Instructions once so every chat matches your tone and format.
Run a Project per topic with its own files and scoped memory.
Build a custom GPT for a workflow you repeat daily, and run Deep Research with a fixed output format.
Frequently asked questions
- Are ChatGPT Custom Instructions worth using?
- Yes. They apply to every new chat, so you stop re-explaining your tone and format. Keep them to concrete rules rather than a long self-description.
- When should I build a custom GPT instead of a Project?
- Build a custom GPT only when the same instructions come up daily and you want a shareable, one-click workflow. For occasional work, a Project with files is simpler.
- What is ChatGPT Deep Research best at?
- It is best for long, sourced reports. Start narrow, add sources iteratively, and give it the final format up front, like a memo or an FAQ.
This is reference content. The prompts and settings run in your own AI tool, not on glunty, so reading this page makes zero requests to those services. Details change often, so verify current pricing and limits on the official site.
Share your result
Made something you like with ChatGPT? Frame the screenshot or keyframe in Beautify a Screenshot: padding, rounded corners, and a soft shadow, then post it. Shrink it first with Compress Image if you need a smaller file. It all runs in your browser.
Official reference: ChatGPT docs. Details change, so verify version-specific features there.
Related on glunty: AI prompt library, Prompting basics, Claude tips.
More AI tool guides
Midjourney parameter cheat sheet
A searchable Midjourney cheat sheet: core parameters like --ar, --sref, --cref, --stylize, --chaos, and --no, each with what it does and an example. Free.
Suno AI tips and prompt cheat sheet
A searchable Suno cheat sheet: section tags, vocal delivery cues, the 5-part style formula, and hacks to stop wasting your daily credits. Free, copy-paste.
Leonardo.Ai token costs and cheat sheet
Learn how Leonardo.Ai tokens are spent per generation and stretch your free daily tokens with this cheat sheet.
Kling AI prompt syntax and cheat sheet
Master Kling AI prompt syntax, storyboard shots, and elements to make consistent AI video within your free daily credits.
ElevenLabs tips and voice settings cheat sheet
Tune ElevenLabs voice settings for narration or character work with a cheat sheet of stability, similarity, and Voice Design.
Ideogram tips and text-in-image cheat sheet
Get reliable in-image text from Ideogram with a cheat sheet on quoting words, Magic Prompt, fonts, and JSON prompting.
Cursor rules tips and cheat sheet
Steer Cursor with rule files and @ mentions: a cheat sheet for .mdc rules, AGENTS.md, globs, and rule precedence.
NotebookLM tips and Audio Overview guide
Get better NotebookLM answers and podcasts: a cheat sheet for source filters, the Customize Audio Overview brief, and Gemini.
Claude hacks for power users
Master Claude with Projects, Styles, and Artifacts for nuanced writing and long-document work.
Runway prompts and tips
Camera moves, Act-One, and Motion Brush prompts to control Runway Gen-4 AI video generations.
Perplexity tips and focus modes
Focus modes, Collections, and Pro Search tips to get sourced answers from Perplexity AI search.
Imagine.art tips and credit hacks
Model picker, credit hacks, and suite workflow tips for Imagine.art image, video, and voice.
Gemini tips and prompt cheat sheet
A searchable Gemini cheat sheet: Gems, Nano Banana image tiers, Deep Research, and grounding, plus hacks that save your Pro quota. Free.
Udio AI music tips and cheat sheet
Master Udio AI music: a cheat sheet for Styles, Style Blend, Sessions waveform editing, Voices, and the instrumental break fix. Free.
Also useful: the AI prompt library and the developer cheat sheets.