Prompt Chain Planner for Complex Tasks
Split a big job into reliable steps.
The Open buttons send the prompt text as currently shown below to that site only when you click. Fill in the placeholders first.
Plan a prompt chain for a complex task. The task and the final deliverable: [TASK_AND_DELIVERABLE]. Input materials: [INPUT_MATERIALS]. Quality criteria: [QUALITY_CRITERIA]. Where mistakes would be costly: [HIGH_RISK_POINTS]. Decide with this tree: IF the task has a single clear output and small input THEN one prompt is enough. IF it needs different skills (extract, analyze, draft, check) THEN use a chain. IF intermediate results must be verified by a person THEN add checkpoints. Return a table: Step | Purpose | Input | Prompt (full text with [PLACEHOLDERS]) | Output format | Check before moving on | Failure handling. Include typical steps: (1) extract facts from sources with quotes and locations; (2) organize into an outline; (3) draft; (4) critique against the quality criteria; (5) revise; (6) final verification list of claims to check. For each step, specify what is passed forward (only the structured output, not the whole conversation) to avoid drift. Then provide: a diagram in text; where to insert human review; how to log each step's output; a test on three sample inputs; how to measure whether the chain beats a single prompt (compare outputs blind with a rubric); the extra cost and time of the chain; and a simplification if it is overbuilt. Mark claims the final answer needs verification against sources.
Fill in before use
Optional: type your details and the prompt above updates. Nothing is saved or sent.
Replace each bracketed placeholder with your own details, then paste into Any AI or another assistant.