19 free prompts · all verified
ChatGPT prompts
The highest-volume prompt search there is — these are structured, reusable ChatGPT prompts with the role, context and format constraints spelled out, so they keep working as the underlying model versions change.
Every prompt lists the exact tool and version it was tested against.
Lock in a recurring role so Custom Instructions stop resetting every new chatWrites the two Custom Instructions fields as a standing role profile for one recurring job — decision rights, refusal rules, and a default output shape — so a brand-new chat opens already calibrated instead of drifting back to ChatGPT's generic default voice.ChatGPT (GPT-5.1)2026-07-22Write a Custom GPT's persona and instructions so it holds up past the first five messagesProduces the Instructions field, knowledge-file guidance, and conversation starters for a Custom GPT other people will actually use without you in the room to correct it — with explicit boundaries on what it should refuse and hand back to a human.ChatGPT (Custom GPTs)2026-07-24Turn a plain-English API description into an Actions schema a Custom GPT can actually callProduces a valid OpenAPI schema, the matching authentication configuration, and call-safety guardrails a GPT Action needs — from a description of an internal API — so the GPT calls the endpoint safely, not just successfully.ChatGPT (Custom GPTs)ChatGPT (GPT Actions)2026-07-25Audit and correct what ChatGPT has actually saved to MemoryA structured pass for reviewing ChatGPT's Saved Memories against what's actually still true, flagging stale, wrong, or overly specific entries instead of a blanket "clear all memory" that throws out the useful ones with the outdated ones.ChatGPT (Memory)2026-07-27Use Canvas for section-scoped edits instead of regenerating the whole documentDirects ChatGPT's Canvas to revise one specific span of a long document in place, with edit-tracking discipline, instead of the default failure mode of rewriting the entire piece every time one paragraph needs a fix.ChatGPT (Canvas)2026-07-26Run an iterative code review loop in Canvas without losing earlier fixesStructures Canvas's code mode as a repeatable review-fix-verify loop with inline suggestions, so successive rounds build on the last accepted state instead of silently reintroducing an already-fixed issue.ChatGPT (Canvas)2026-07-28Give Deep Research a mission brief precise enough to trust the citationsStructures a Deep Research task with an explicit scope boundary, source-quality bar, and deliverable shape, so the agent spends its research budget on the actual open question instead of drifting into adjacent, easier-to-answer territory.ChatGPT (Deep Research)2026-07-29Set up a Project so every chat in a workstream shares the same ground truthConfigures a ChatGPT Project's files and project-level instructions as durable shared context for an ongoing piece of work, with an explicit policy for what belongs at the project level versus what stays in one thread.ChatGPT (Projects)2026-07-30Structure an Advanced Voice Mode session for language practice that actually corrects youSets explicit correction rules, pacing, and code-switching boundaries for a spoken practice conversation in Advanced Voice Mode, so the session doesn't default to the polite-conversation-partner behavior that lets every mistake slide to keep the conversation flowing.ChatGPT (Advanced Voice Mode)2026-07-21Configure a Scheduled Task with a trigger and output contract that won't driftWrites a recurring ChatGPT Task's instructions with an explicit trigger condition, source list, and output shape, so a weekly automated digest stays consistent in format run after run instead of quietly reinterpreting the brief differently each week.ChatGPT (Tasks)2026-07-23Audit an uploaded dataset in Code Interpreter before it produces a single chartForces a data-quality pass — types, nulls, duplicates, outliers, and join integrity — on an uploaded file before Code Interpreter runs any analysis or chart request on it, so a confident-looking chart doesn't quietly rest on a column full of silently miscoerced values.ChatGPT (Code Interpreter)ChatGPT (Data Analysis)2026-07-24Reconcile a stack of receipt photos against an expense policy in one passUses Vision to extract structured line-item data from multiple receipt photos at once and checks each one against a stated expense policy, flagging exactly which line and rule triggered a compliance issue instead of a flat pass/fail per receipt.ChatGPT (Vision)2026-07-27Structure a live web-search query so it separates verified facts from best guessesDirects ChatGPT's web search on a fast-moving or recent topic to tag every claim with a confidence level and a source, rather than blending confirmed facts and plausible inference into one uniformly confident-sounding answer.ChatGPT (Web Search)2026-08-01Get ChatGPT's chat interface to emit JSON reliable enough to feed into a scriptBuilds a JSON-output contract for the consumer ChatGPT app — where there is no schema-enforcement toggle the way the API has — with an explicit self-check step, since the chat interface's adherence depends entirely on how rigorously the prompt itself specifies and validates the shape.ChatGPT (GPT-5.1)2026-07-31Write a GPT Store listing that gets picked over the ten other GPTs like itProduces the public-facing name, description, category, and conversation starters for publishing a Custom GPT to the GPT Store, written to differentiate a specific real capability from the generic competing listings a searcher will see next to it.ChatGPT (GPT Store)2026-07-29Structure a sensitive one-off request as a Temporary Chat, deliberatelyFrames a genuinely sensitive, one-time request for a Temporary Chat session — which doesn't use Memory, doesn't appear in history, and isn't referenced by future chats — with the context normally supplied by Custom Instructions and Memory restated explicitly, since none of that ambient context carries into this mode.ChatGPT (Temporary Chat)2026-07-20Get a structured comparison across several uploaded PDFs, not a summary of each oneDirects ChatGPT to build a cross-document comparison table across multiple uploaded PDFs — flagging contradictions and coverage gaps between them — instead of the default failure mode of producing one separate summary per file that leaves the actual comparing to the reader.ChatGPT (GPT-5.1)2026-08-02Pick the right reasoning depth for the task before you type the promptA short decision pass for choosing between ChatGPT's fast/instant response mode and its extended-thinking mode for a specific task, plus how to phrase the request so it actually invokes the depth chosen instead of getting auto-routed to the wrong one.ChatGPT (GPT-5.1)2026-07-30Build a weighted decision matrix that can't be gamed by picking favorite-friendly weightsConstructs a weighted-criteria decision matrix for choosing between real options, with an explicit sensitivity check showing whether the recommended winner actually holds up across a reasonable range of weightings, instead of a single static score that quietly depended on one specific weight split.ChatGPT (GPT-5.1)2026-07-26
