Every IEP progress report drafted from cited goal data, with the case manager signing every goal
Every goal rated on at least three cited data points, signed in an afternoon.
See one case, screen by screen ↓It drafts every IEP progress report from the goal data itself. Seven agents pull each probe, gradebook score and provider note onto the goal's chart with its source, refuse to rate a goal on fewer than three points, apply the district's decision rule, and write each narrative in the family's language beside the English. The case manager reviews and signs every report.
AI in education: a buyer's guide to agentic AI for colleges, universities and school districts →
Updated
An IEP progress report tells the family, in writing and at least as often as report cards, how the student is progressing toward each annual goal (34 CFR 300.320(a)(3)). For a case manager with 24 students and 81 goals, that means opening the progress monitoring assessment platform for probe scores, the gradebook for writing samples and the IEP system for provider notes, copying each number by hand and losing its probe and date on the way, then judging each goal by eye and typing a paragraph per goal.
When data is thin, the report says "progressing" anyway or repeats last period's code. When four points in a row fall below the aimline, the district's decision rule says convene the team, but nobody applies it to 22,780 goals a period. Families who read Vietnamese or Spanish get English. And when a state complaint arrives, which probe, which date, who edited the narrative and what the family was sent lives in four places. IEP Progress Monitoring & Progress Reports puts seven agents in front of that work, so the case manager reviews and signs instead of copying and guessing.
Estimated split for one case manager's 81 goals, by hand and with the agents; the 'before' column is about 12 minutes a goal, the 'after' column is about three hours, one afternoon. Our arithmetic, kept modest.
Seven agents pull, check, chart and draft every goal; the case manager signs, the liaison checks the translation, the coordinator samples.
Lina K. is in grade 6 at Linden Avenue Middle School in Marlow Bay Unified. Her IEP has four goals, her parents read Vietnamese, and period 1 reports go home with report cards on Friday, October 30. Here is her report, screen by screen, in the working solution.
Joanna's view opens on the period's path: data cutoff Oct 16, late data closes Oct 23, sign by Oct 27, the coordinator sample Oct 28–29, home with report cards Oct 30. The agents' brief says all 81 goals on her caseload are drafted from cited data points, and Needs you lists four data requests, a Spanish question about Mateo's reading and five reports ready to sign.
Each goal is a row: a small chart with the baseline, the dashed aimline and every probe with its value, then the drafted narrative for the family, then the rating with the agent's reason and confidence. Goal 1, reading fluency, is on track: seven points from 80 to 86 words correct per minute, projected about 116 of 115, confidence 0.97.
4 goals, 19 data points, each cited.
Point 4 of 7 on Goal 2 is Mixed computation, grade 5, form 4, given by Ari Brennan on Wednesday, September 23: 16 correct digits in 2 minutes against an aimline of 16.4 that day. The Data agent read it from the progress monitoring assessment platform and matched it exactly to Goal 2's measure.
Goal 2's seven scores run 15, 16, 17, 16, 16, 17, 17. The aimline needs 0.72 correct digits a week; the trend is 0.25, which projects about 24 of 40 by May. The last four fell below the line, so a plain band lights across the row and the Trend agent cites the district's rule 4.2.
Rule 4.2 tripped: 4 in a row below the aimline · recommends an IEP team meeting · the case manager decides.
The recommendation quotes the rule, lists the four scores with the aimline value for each date, and checks the services: 24 of 25 sessions delivered, so the plan, not attendance, is the question. Joanna has three choices: propose a team meeting, change the intervention first, or not now, the last two with a reason.
Joanna proposes the meeting. The agents draft the participants under 34 CFR 300.321 (both parents, Joanna, the math 6 teacher, Ari Brennan and the coordinator), Linh Tran as Vietnamese interpreter under 300.322(e), three times in Room 114 and an agenda. Nothing goes out until Linh checks the Vietnamese and sends it herself.
Mai Kiều replies in Vietnamese, and the school reads it beside its English translation at confidence 0.97. Joanna accepts the time in one click.
"Thank you, Ms. Reyes. Thursday, November 5 at 3:30 p.m. works for us. My husband will come too. Does Lina need to come with us?"
Goal 3, written expression, had two writing probes, and the district rates a goal only on three or more, so the Sufficiency agent wrote no rating and no narrative and asked Hannah for one. Her own view shows only the goals she measures, never the whole IEP or the disability category. She enters the Oct 16 score in one step and Goal 3 can be rated.
Joanna reviews the four goals and signs. The report now has 20 data points, each cited, and none without enough data; Goal 2's band reads "meeting set for Thu, Nov 5, 3:30 p.m." Linh reads each Vietnamese narrative beside the English before the report goes home.
The sample at Linden Avenue holds 24 of 126 reports: 11 always, because a rule tripped or a rating was changed, and 13 drawn at random. The Sample agent pre-checks six things on each; on Owen M.'s report the narrative says "from 12 to 18" while the data shows 12 to 16, and Gwen returns it with a reason.
The same report, seen by the four people who carry it: what their report week looked like, and what it looks like now.
Seven agents pull, check, chart and draft every goal; the case manager signs, the liaison checks the translation, the coordinator samples.
Reads probe scores, gradebook scores and provider notes and matches each to the goal whose measure it is, keeping the probe, form, date and who gave it; leaves out a score whose level or form does not match, and says why.
Counts each goal's points; below three, writes no rating and no narrative, asks the person who holds the data in one short message, and tells the case manager after three school days.
Draws the aimline, fits the trend at five or more points, applies the district's decision rule, checks the services were delivered, projects to May and recommends an IEP team meeting or a goal review.
Writes three to five sentences per rated goal at a grade 6 reading level, with every number taken from a cited point.
Writes the family's language beside the English on reports, invitations, replies and confirmations.
Draws the coordinator's sample, every tripped rule and changed rating plus a random 10%, and pre-checks six things on each report, including whether the narrative's numbers match the data.
Builds a records pack for a state complaint, due process hearing, monitoring visit or family request, redacts other students and seals it with SHA-256.
Sign, check and sample. The agents propose; a named person decides.
The case manager can ask about her caseload in plain words; every answer cites the probe, the rule or the message it comes from, and opens the screen to act on.
Which goals on my caseload are not on track?
Three goals tripped the district's decision rule: Lina K., Goal 2, math computation, 4 below, meeting proposed; Mateo R., Goal 1, reading comprehension, 4 below, intervention changed Oct 15; Elijah F., Goal 3, on-task behavior, 5 below, meeting Oct 28.
Goal 2: how is Lina doing?
Not on track. 7 points from Wed, Sep 2 to Wed, Oct 14: 15, 16, 17, 16, 16, 17, 17. 4 points in a row below the aimline; trend 0.25 a week, aimline needs 0.72. Projected by May: about 24 of 40. Rule 4.2 tripped; I recommend a team meeting, and you decide.
Why is Goal 3 not rated for Lina?
Goal 3 has 2 writing probes this period, Sep 11 (19) and Oct 9 (21), and the district rates a goal only on 3 or more. Her Oct 16 quick-write has a rubric score, which is a different measure. I asked Hannah Brooks on Oct 19 to count its correct writing sequences; it is due tomorrow.
What did Lina's parents say?
Mai Kiều wrote in Vietnamese. In English: "Thank you, Ms. Reyes. Thursday, November 5 at 3:30 p.m. works for us. My husband will come too. Does Lina need to come with us?" You can accept the date from the thread.
12 screens from the working solution, on its sample data. Pick one to see it large.
The rules every IEP progress report is drafted against are settings, versioned, and every change is on the audit trail.
| Setting | Default | Choose from |
|---|---|---|
| Reporting periods | 4 a year: Oct 30, Jan 29, Apr 9, Jun 11 | 3 a year, with trimester report cards |
| Fewest data points to rate a goal | 3 points | Below this the agents ask for data instead of rating |
| Points needed to project to May | 5 points | A trend on fewer points is not shown |
| Decision rule: in a row below the aimline | 4 points | Then recommend an IEP team meeting (rule 4.2) |
| Data requests answered within | 3 school days | Then the case manager; the coordinator after 2 more days |
| Coordinator sample | 10% of signed reports, at random | Plus, always, every report where a rule tripped or a rating was changed |
| Send to a professional translator below | 0.90 agent confidence | 0.85 or 0.95 |
| Keep records | 5 years after the student leaves | 6 or 7 years; the family is told before destruction (300.624) |
Every figure is labelled: a target the solution is built to, an estimate, a typical published result, or a proven one.
Sources: 34 CFR 300.320(a)(3), 300.321, 300.322(e), 300.324(b), 300.503(c), 300.610–300.626 and 300.613 · Council for Exceptional Children, analysis of 2024 IDEA data (8,194,424 students ages 3–21 served under Part B, up 3.8% in a year) · decision rules follow common data-based individualization practice, as the National Center on Intensive Intervention describes; each district sets its own · "target" = design goal, measured in the live solution · "demo" = seen in the working solution, on its sample data · "typical" = published figures · "estimated" = our arithmetic, about 12 minutes a goal by hand. Marlow Bay Unified School District, its rule 4.2 and the people named on this page are fictional: characters and sample data in the working solution.
It drafts every IEP progress report from the goal data itself. Seven agents pull each probe, gradebook score and provider note onto the goal's chart with its source, refuse to rate a goal on fewer than three points, apply the district's decision rule, and write each narrative in the family's language beside the English. The case manager reviews and signs every report.
Under 34 CFR 300.320(a)(3), the IEP says when periodic reports on progress toward the annual goals are provided, for example at the same time as report cards. In the sample district that is four times a year, and the reporting periods and dates are a setting.
The Trend agent draws the aimline from the baseline to the goal, fits a trend once there are five or more points, and applies the district's decision rule: four points in a row below the aimline means it recommends an IEP team meeting. It also checks the services were delivered. The case manager decides, and can change any rating with a reason code.
Below three points the Sufficiency agent writes no rating and no narrative. It asks the teacher or provider who holds the data, starts a three-school-day clock, and then tells the case manager. A goal can still be reported as not enough data, with a reason, so nothing is guessed.
The Narrative agent writes three to five plain sentences per rated goal at a grade 6 reading level, every number from a cited point, and the Family language agent puts the family's language beside the English. The parent liaison checks every translation and sends it; below 0.90 confidence a professional translator takes it.
Yes. The agents draft, check and recommend; they never sign, rate below the minimum, decide a meeting or send to a family. The case manager signs every report, the parent liaison sends every translated notice, and the coordinator who samples a report cannot be its signer.
Yes. The Records pack agent builds a pack in one step, with every data point and its source, each version and edit, both languages as sent, family messages and the access log, other students redacted and the pack sealed. It is built for family records requests too, which 34 CFR 300.613 says must be met without delay and within 45 days.
The Agentic Solution Engine builds and deploys it from your requirements and documents, such as your procedures manual's decision rules and a sample of past goals and probe data, and it goes live once every quality gate has passed. We will walk you through it on your own data first.
Every draft IEP read against the rule before the meeting, not a sample after it.
1-min demoSpecial Education Due Process & State ComplaintsChronologies in hours, not weeks, with every allegation tied to the page that proves it.
1-min demoSection 504 Plan ManagementEvery teacher sees the plan, and every test is checked before it starts.
1-min demoWe'll run IEP Progress Monitoring & Progress Reports on a sample of your own goals and probe data.