TextIntermediate

Audit an automation you have not looked at in six months

You are reviewing a workflow that has been running unattended. Below are its steps, triggers, tools and prompts, any run logs I have, and what it was built to do.

1. STILL TRUE? - table: Original assumption | Evidence it still holds | Verdict (holds / broken / unknown). Cover data shape, volume, tool behaviour and who reads it.
2. SILENT FAILURES - where this can produce a wrong or empty result without erroring, and the check that would catch each one.
3. DEAD WEIGHT - steps, branches or tools that no longer do anything, and what confirms it before deleting.
4. DRIFT RISKS - what changed outside the workflow (renamed fields, new pricing, model deprecations, people who left) that it depends on.
5. FIX LIST - ordered, with effort (minutes/hours/days) and what breaks if I skip it.
6. RETIRE OR KEEP - one recommendation, including whether anyone still uses the output.

Rules: never assume behaviour of a tool you were not shown - write "verify in the tool". No rebuild suggestion unless the fix list is longer than the rebuild.

WHAT IT DOES AND WHO USES IT: {{paragraph}}
STEPS, TRIGGERS, PROMPTS: {{paste export or describe each node}}
RUN LOGS / ERRORS / VOLUMES: {{paste, or "none"}}

How to use it

Paste the actual node export and any run history - without volumes the silent-failure section stays generic. It cannot execute your workflow, so every "verify in the tool" line is a real task for you, not a formality.

Compatible popular AI tools

These tools are mapped to this prompt based on their capabilities.