What the Spend Request Agent does
It normalizes incoming spend requests and runs your policy and evidence checks: request, amount, policy check, evidence, approver and status. Approvals become one review, not a reconstruction.
Unrecorded approvals are next year's audit problem. This assistant structures each request as it arrives, checks it against your written policy, lists missing evidence, and drafts the response - with the decision trail saved per request. Hallway asks become a queue you can actually clear.
Where the Included files go in ChatGPT Dots
Give skill.md, setup-prompt.txt and usage-prompt.txt to your dot as reference instructions it can read. Ask it to restate the job and output contract, then run the setup prompt on fictional data. Review the result before connecting real records. Text instructions do not install a plugin or grant access. OpenAI says local skills require a connected computer, online with the ChatGPT app open.
What a run looks like on ChatGPT Dots
Ask your dot to run the Spend Request Agent with the approved inputs. It should return the workflow's evidence and unknowns for review, not act on missing facts. Review the saved result and proposed actions before anything is sent, spent, published or scheduled.
Setup, step by step
- Check eligible Dots access and connect only the source account this job needs.
- Provide skill.md, setup-prompt.txt and usage-prompt.txt as readable reference instructions. A local skill needs a connected computer.
- Run the fictional setup test and inspect the actual output.
- Use the usage prompt with authorized inputs for a preparation-only real run. Review the destination and saved result before approving actions.
Fictional test: A fictional request has an invoice but no approved policy threshold. Expected: threshold unknown and approver question, not approval or payment.
Honest limits on ChatGPT Dots
Dots access is rolling out gradually on eligible plans and accounts. Create your dot on desktop first; mobile web is not supported. Plugins, messaging channels and computer access are separate. Check the connected account and permitted actions. Built-in safeguards and automatic action review apply; custom rules do not override them. This is a documentation-based guide, not a live Dots account test.
Questions people ask
Does the Spend Request Agent work with ChatGPT Dots?
These are documentation-based setup notes for a readable text workflow. Verify your access, plugin actions and sample output first; no live Dots account test is claimed.
What should I review?
The source evidence, unknowns, output file and every proposed external action. The included instructions do not grant access or approval.
Is this affiliated with OpenAI?
No. AI Agent Skills is independent of OpenAI.
Run this agent on another assistant
Microsoft Copilot Hermes Instinct OpenClaw Muse Perplexity ChatGPT Claude DeepSeek Town Grok Gemini
More finance teams agents on ChatGPT Dots
Cash Flow Forecast Agent Budget Variance Agent Management Reporting Agent
AI Agent Skills is an independent store, not affiliated with ChatGPT Dots or its makers. Platform names belong to their owners.
Checked September 30, 2026 by the AI Agent Skills team. Independent of OpenAI. Exact assistant-by-job monthly search volume is unverified. Sources: Meet dots and access, Computers and apps, Controls, OpenAI safety FAQ.