# How do I structure prompts so Cowork works correctly and reliably?

Source: https://zilliz.com/ai-faq/how-do-i-structure-prompts-so-cowork-works-correctly-and-reliably
Author: zilliz

To make Cowork reliable, structure prompts like a short engineering spec: **goal, scope, constraints, outputs, validation, and escalation rules**. Start with a single-sentence goal that is measurable (“Create `out/report.md` summarizing all docs and extract decisions into `out/decisions.csv`”). Then define scope precisely (“Only use files under `/Shared/ProjectDocs`; ignore `tmp/` and `node_modules/`; process only `.md`, `.txt`, `.pdf`”). Next define constraints that prevent surprise behavior: “Do not delete anything,” “Do not overwrite originals,” “Write all new files to `out/`,” “If a file is unreadable, list it in `out/errors.csv`,” and “Do not use the internet unless I explicitly ask.” When people say Cowork “did the wrong thing,” it’s usually because the prompt omitted one of these: scope, constraints, or concrete deliverables.

For complex tasks, force checkpoints and make ambiguity handling explicit. A repeatable pattern is:
1) **Inventory**: write `out/inventory.csv` with file paths, sizes, modified times, and detected type.
2) **Plan**: write `out/plan.md` listing every operation (rename/move/edit) and why, then wait for confirmation.
3) **Execute**: apply changes, but log every action to `out/actions.log`.
4) **Validate**: run sanity checks (counts, schema, missing fields) and write `out/validation.md`.
5) **Deliver**: write final outputs plus a “what changed / what didn’t” summary.
Also specify escalation rules: “If you’re not confident about a classification, label it `unknown` and add it to `out/open_questions.md`.” “If two sources conflict, prefer the newest timestamp and record both in `out/conflicts.csv`.” These rules keep Cowork from “filling in” uncertain details, which is a common failure mode in agent workflows.

If your task feeds a larger workflow, bake the downstream contract into the prompt so Cowork outputs are immediately ingestible. For example: “Chunk documents into 400–800 token sections,” “Assign stable IDs like `DOC-###_S##`,” “Emit `out/metadata.jsonl` with fields `doc_id, chunk_id, title, source_path, updated_at, tags`,” and “Write chunk files into `out/chunks/`.” That makes it trivial to validate and index into a vector database such as [Milvus](https://milvus.io/) or [Zilliz Cloud](https://zilliz.com/cloud). The more your prompt looks like a schema + acceptance test, the more Cowork behaves like a reliable batch tool instead of a “helpful but improvisational” assistant.