Progress reports

A learner’s attendance, scores, study time and progress, for the AI app to write their report.

Each tool below is one switch in Settings → MCP Server. Your AI app sees a tool only when it is switched on for your role; every call is checked again on the server.

learner_report

Progress reports: view data · Read only

Let the connected AI app read the data behind a learner’s progress report — attendance and every live class, test scores with rank and class average, marks by subject, study time and what was studied, course progress next to the batch average, assignments with feedback, doubts with their replies and the analysis of each attempt — plus the reports already generated in the dashboard, so it can write the report for you. Read-only. The AI app writes the report itself, so it uses no AI credits.

Action	What it does
guide	what every field means, the dashboard report’s status/grade rules and how to write a grounded report. Read it once before writing.
data	report = the facts the dashboard report is built from (attendance, live classes, assessments with rank/percentile/class average, marks by subject, study habits, course progress, certificates, assignments, doubts, logins, topic mastery / Bloom’s / confidence / misconceptions); detail = what that report throws away (every live class, per-module progress next to the batch average, what was studied slide by slide, assignment submissions with feedback, the doubts’ text and replies, and the analysis of each attempt). Ask for only the modules / detail the admin’s report needs when the window is long.
saved_reports	progress reports already generated in the dashboard (to compare periods).
saved_report	one of them in full.

Parameters

action · string · required

The action to run: guide, data, saved_reports, saved_report.

learner_id · string

The learner’s learner_id (from learners find).

start_date · string

data: window start, YYYY-MM-DD.

end_date · string

data: window end, YYYY-MM-DD (inclusive).

batch_id · string

data: one of the learner’s batch_ids (from learners get) to scope progress / assignments. Default: their latest batch.

modules · string[]

data: report sections to include. Default: all.

detail · string[]

data: detail parts to include. Default: all but attempt_details (the per-question input of each attempt — large). [‘none’] = no detail.

attempt_limit · integer

data: newest attempts to include in attempt_insights (default 10, max 50; attempt_details max 15).

report_id · string

saved_report: a report_id from saved_reports.

page · integer

saved_reports: page, from 0.

What your AI app is told about this tool

A learner's progress-report data, for YOU to write the report (for the admin, a parent or the learner). Read-only; uses no AI credits — you are the writer. Pick an `action`:
- guide: what every field means, the dashboard report's status/grade rules and how to write a grounded report. Read it once before writing.
- data (learner_id, start_date, end_date — YYYY-MM-DD, default the last 30 days, max 366 days; optional batch_id, modules, detail, attempt_limit): `report` = the facts the dashboard report is built from (attendance, live classes, assessments with rank/percentile/class average, marks by subject, study habits, course progress, certificates, assignments, doubts, logins, topic mastery / Bloom's / confidence / misconceptions); `detail` = what that report throws away (every live class, per-module progress next to the batch average, what was studied slide by slide, assignment submissions with feedback, the doubts' text and replies, and the analysis of each attempt). Ask for only the modules / detail the admin's report needs when the window is long.
- saved_reports (learner_id): progress reports already generated in the dashboard (to compare periods).
- saved_report (report_id): one of them in full.
Get the learner_id from learners(action='find'). Never invent numbers: every figure you write must come from this data.