Complete AI Training
Sign inGet my AI kit

Your job's AI kit

Get your AI kit

Tell us who you are and what you do. We show you your kit right away and email you the link: skills, prompts, AI agents, MCP servers and courses for your job.

500+ jobs ready, and we make a kit for any other job. No payment needed to look.

Share

AI agent for teaching assistants

Learner Project Milestone Agent

Learners make real, rubric-backed progress on their projects.

Learner Project Milestone Agent: what goes in, what the agent does and what you get

What it does

Learners in project-based courses often stall with vague next steps like 'keep working on it'. This agent compares each uploaded piece of work with the milestone rubric and finds the biggest gap. It proposes one feasible next learning task aimed at that gap, sized to the learner's available time. When the learner uploads new work, it checks it against the milestone criterion again. If the criterion is still not met, it refines the task and tries again. Activity alone, such as many edits or hours logged, never counts as progress. It records milestone progress in a record the learner owns, and the instructor can review it. Edge case: work that meets the criterion in an unexpected way is accepted.

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.

Start and resultWhat it doesA check on its own workWaits for your OKGoes back and retries
Yes, continueNo 1 STARTS WHEN Artifact uploaded 2 USES A TOOL Compare the artifact with the milestone rubric 3 DOES Propose a feasible next task for the biggest gap 4 USES A TOOL Check the returned artifact 5 CHECKS THE RESULT Does it meet the milestone criterion? If not: refine the task. Back to step 3. 6 DOES Record milestone progress 7 RESULT Learner-owned progress record
Read the steps as a list
  1. Artifact uploaded
  2. Compare the artifact with the milestone rubric
  3. Propose a feasible next task for the biggest gap
  4. Check the returned artifact
  5. Does it meet the milestone criterion?If not: refine the task. Back to step 3.
  6. Record milestone progress
  7. Learner-owned progress record

How it decides

It picks the next step from the gap between the artifact and the criterion and refines the task when the result misses it.

  • Next task: aimed at the biggest rubric gap.
  • Progress: counted only when the milestone criterion is met, not for effort.
  • Task size: fits the learner's stated available time.
  • Refine: make the task narrower after each failed check.

Make it yours

Every agent is a starting point. You choose these settings for your own situation.

  • Milestone rubric per project
  • Task size limit (default 2 hours of work)
  • Whether the instructor sees each step (default weekly summary only)
  • Tone of task wording (default direct and encouraging)
  • Number of refinements before the instructor is asked (default 3)

What keeps you in control

It always asks you first

  • Final grades
  • Doing assessed work for the learner

Hard limits

  • Never completes assessed work for learners.

It stops when

  • Done: milestone met.

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.

10 minto set it up in your AI
5 AIsChatGPT, Claude, Copilot, Gemini, Grok
  • 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
Get access to this agent

An example run

What happensOn April 14 learner Dev Malhotra uploads a survey report draft. It summarizes counts but does not answer the research question. The agent sets one task: state the answer in a paragraph with two numbers as evidence. The returned paragraph names the answer but uses no numbers, so the check fails. The agent narrows the task to choosing two supporting figures. The next version meets the milestone.

More agents for teaching assistants