01 / Prepare
Bring a little context.
Use a text chat assistant you already have. No paid feature is required for the prompt; usage and message-length limits may apply. Use only information you have permission to share.
- The recurring task as you perform it today.
- Trigger or schedule with a time zone.
- Systems, fields, and one example input/output.
- Decisions that still need a person.
02 / Do the work
From input to something useful.
- 01
Describe a complete run
Include awkward exceptions and who handles them today.
- 02
Create the specification
Give the assistant the prompt and a representative, anonymized example.
- 03
Walk through the edge cases
Try normal, missing-data, and duplicate inputs on paper. Fix unclear rules.
- 04
Verify before connecting
Check your chosen automation tool’s integrations, permissions, and pricing. This guide creates a plan, not a running automation.
03 / Make it yours
Build your prompt.
Fill in the blanks and paste your material. The prompt updates as you type. Copy it into your assistant; the checks below are part of the workflow.
What you type stays in this page. Nothing is sent to Deadbounce or saved; close or reload the page and it’s gone.
Turn this recurring task into an automation specification. Task: [description] Trigger/schedule/time zone: [details] Current tools: [tools] Inputs and required fields: [fields] Desired output: [details] Human decisions: [details] Representative input and expected result: [example] Produce a specification, not a live automation. Do not claim connections or schedules have been created. Include: purpose and success condition; trigger; required inputs; numbered flow separating fixed rules from AI interpretation; missing-data and failure behavior; proposed duplicate prevention; human review points; notification recipient; manual fallback; access, integration, and pricing questions to verify. Mark unconfirmed choices “Proposed.” Do not assume features exist or integrations are free. Finish with three sample cases: normal, missing required information, duplicate input. State expected behavior and what to inspect for each. Prefer the smallest useful version.
04 / What this looks like
A small example.
Illustrative input and output. This is not a measured result or a promise of what a model will return.
Every Monday at 9 a.m. America/New_York, turn the project table into a draft weekly update for the team lead.
Trigger: Monday, 9 a.m., America/New_York. Action: Draft update from current rows. Missing status: Flag project for review. Proposed duplicate rule: One draft per reporting week. Review: Team lead approves before sharing.
05 / The important part
Check before you use it.
Work from the original material, not from how confident the draft sounds.
Tick each check against the source, not against how confident the draft sounds.
Automating only the happy path leaves someone cleaning up duplicates and failed runs.
Original Deadbounce workflow · Published September 18, 2026.
Tool requirements checked against official documentation on September 25, 2026. These sources describe the tools; they do not endorse this workflow. Suggested times are estimates, not benchmarks.
- Claude: getting started and free-plan limits
- ChatGPT: what the free tier includes
- Gemini: using text prompts and reviewing responses
- Gemini: usage limits with and without a paid plan
- Companion tool and plan details