classdojo-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| CLASSDOJO_CDP_URL | No | URL of the Chrome DevTools Protocol endpoint used to connect to the local browser adapter (e.g., http://127.0.0.1:9222). |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
| prompts | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| classdojo_list_classesA | Lists the three-digit classes visible in the locally connected ClassDojo teacher session. |
| classdojo_inspect_workbookA | Scans every worksheet and reports candidate class, seat-number, and student-name column groups without importing or transmitting student data. |
| classdojo_get_rosterC | Reads a class roster without changing ClassDojo data. |
| classdojo_get_skillsA | Reads positive and needs-work skills, including points and icons, without changing ClassDojo data. |
| classdojo_get_ui_stateA | Reports whether the selected class is ready, adding students, or blocked by a family-invitation, welcome-message, or student-account dialog. It does not dismiss anything. |
| classdojo_preview_roster_importB | Reads a local XLSX workbook and reports the student-name changes before anything is sent to ClassDojo. |
| classdojo_apply_roster_importA | Sends student names to ClassDojo only after explicit confirmation, then verifies each class by reading it back. |
| classdojo_verify_roster_against_workbookA | Reads the current ClassDojo roster and a local XLSX workbook, then reports exact count, missing, and unexpected student-name differences without changing ClassDojo data. |
| classdojo_preview_skill_syncA | Compares a preset, source class, or inline rule list with target classes before any skill is changed. |
| classdojo_apply_skill_syncA | Exactly replaces target skills only after explicit confirmation, then reads every class back. |
| classdojo_verify_skill_syncA | Compares current target skills with a preset, source class, or inline rule list without changing data. |
| classdojo_doctorA | Checks the local browser connection, login visibility, and available classes without writing data. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
| classdojo_configure_skill_rules | Guides a review-first workflow for the built-in Traditional Chinese classroom skill preset. |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 12 tools
Most tools have clear purposes, but several pairs overlap heavily: preview_skill_sync and verify_skill_sync both compare skills against a source without changing data, and verify_roster_against_workbook and preview_roster_import both report workbook-vs-roster differences. The descriptions help, but an agent could easily select the wrong verification/preview tool.
The classdojo_ prefix and snake_case naming are consistent, and most tools follow a clear verb_object pattern (get_roster, list_classes, apply_skill_sync, preview_roster_import). The main deviation is classdojo_doctor, which is vague and does not follow the verb_noun convention used by the rest of the set.
With 12 tools, the server stays within a well-scoped range for its domain. Each tool supports a distinct step in the roster or skill-sync workflow, including inspect, preview, apply, and verify phases, so none feel redundant.
The tool set covers the full roster-import and skill-sync lifecycle: inspect/preview/apply/verify for both workflows, plus connection diagnostics and UI-state checks. Minor gaps exist, such as no way to dismiss or resolve UI blockers and no explicit rollback tool, but these are workable limitations.