AI agent for elected officials
Campaign Promise Delivery Tracker Agent
A sourced status table for every promise that the official can publish with confidence
What it does
After an election, promises fade, and nobody can say which have been kept. This agent starts from the campaign or manifesto and lists each promise as a specific, checkable commitment. It then links every promise to the bills, budget lines and executive actions that relate to it. It scores progress using a simple scale, from nothing started to delivered, and lists promises with no action at all. For each claim it records the source, such as a bill number or a budget page. Before drafting the status update, it checks that every claim has a source and that dates and amounts match the documents, and it fixes those that do not. The official approves what is published. Edge case: a promise that was met in part by a program funded below the promised amount is scored as partial with the numbers shown.
How it works
Follow the arrows from top to bottom. The orange dashed arrow is the loop: when a check fails, the agent goes back and tries again.
Read the steps as a list
- Quarterly status review
- Load the promise list and the previous status table
- Search bills, budgets and executive actions for each promise
- Link each promise to evidence and score its progress
- Does each score have a source document?If not: search again for evidence or lower the score to the proof available. Back to step 3.
- List promises with no action
- Draft the status update with changes since last quarter
- Do dates and amounts in the draft match the sources?If not: correct the figures from the documents and redraft. Back to step 6.
- Official approves what is publishedThe agent waits here for your OK.
- Published promise tracker
How it decides
Each promise is scored by evidence: none, proposed, passed, funded, delivered. A claim is kept only if a source document supports it.
- Score a promise as delivered only with evidence of completion
- Score funded below the promised amount as partial and show the figures
- List a promise with no action for 12 months as stalled
- Cite a source for every claim
Make it yours
Every agent is a starting point. You choose these settings for your own situation.
- Promise list and wording
- Scoring scale
- Review frequency
- Publication format
What keeps you in control
It always asks you first
- Publication of the status update
Hard limits
- Never publishes
- Never scores a promise without a source
It stops when
- Done: the status table is approved
- Stop: the promise list is not defined
Set it up
We guide you through the set-up, step by step
Members get the full set-up guide for this agent. No technical skills needed: you copy, paste and upload.
- One set of instructions to paste into your AI, with the clicks for ChatGPT, Claude, Microsoft 365 Copilot, Gemini and Grok
- The agent then walks you through connecting your own data, one source at a time
- A downloadable copy with the flow chart, the rules and the full guide
