Choose the action that respects the task, evidence, and effect boundary. Each answer includes a short reason.
Review the decisions
- Prompt intent: turn an open request into an acceptance check
- Prompt inputs: normalize records before asking for conclusions
- Few-shot prompting: select examples that cover decisions, not just easy cases
- Output contracts: parse a result and preserve an explicit unknown state
- Prompt decomposition: split stages at verifiable handoffs
- Retrieved context: select sufficient evidence before writing the answer
- Evidence IDs: make generated claims auditable against supplied records
- Long context: make inclusion and truncation testable
- Document extraction: separate observed fields from inferred values
- Tool calls: validate intent and arguments before an external effect
- Tool loops: set budgets, state checks, and a stopping condition
- Multimodal prompts: separate what an image shows from what it suggests
Common Mistakes
- Treating readable output as a verified decision.
- Allowing untrusted context to grant an action.
- Ignoring an explicit unknown state.