The prompt
Check these spend requests against our policy: [paste requests, then policy rules]. Per request: pass/fail/unclear per rule, missing evidence, and the approver level it routes to. Do not approve or reject - that decision belongs to the approver.What to bring
Spend requests arrive by email, chat and the occasional hallway ask. Policy checks happen in someone's head, and audit season reconstructs decisions nobody wrote down.
What a good run returns
Request, amount, policy check result, evidence, approver and status, in a Google Sheet in a Finance Spend Request Workspace folder.
Where the free version ends
A prompt runs when you remember to run it, from whatever context you paste, and it forgets the rules between sessions. The Spend Request Agent ($19 one-time) installs this as a standing workflow: fixed evidence rules, a saved output in your own Drive every run, fictional-data testing before real work, and hard stops before anything sends, spends or publishes. More detail on the spend request approvals guide.
Questions people ask
Is this prompt really free?
Yes. Copy it, paste it into any capable assistant, bring the inputs listed. No signup, no email gate.
Which assistants does the prompt work in?
Any capable chat assistant: Instinct, ChatGPT, Claude, Gemini, Grok, DeepSeek, Perplexity, Microsoft Copilot, Hermes, OpenClaw, Town, Muse and others.
What is the difference between the free prompt and the $19 agent?
The prompt is a one-off: you re-paste the rules every run. The $19 agent is the same job productized - tested setup, saved files, consistent output, standing cycle.
Last updated 13 September 2026 · Reviewed by the AI Agent Skills team.