Claude MCP skill for Schoology — access all your school data from Claude
14 tools with READ_ONLY operations. All tools have basic descriptions (10-50 chars), but lack depth for LLM selection guidance. Input schemas visible for 6/14 tools; 8 tools have no documented input parameters despite accepting them (get_me, list_my_sections, get_my_grades, list_assignments, list_inbox, list_sent_messages, list_my_events, list_discussions). No output schemas documented. No error handling guidance. No parameter descriptions. Tool names follow verb_noun pattern (good), but descriptions are too generic to guide agent behavior. No pagination hints despite list_* tools. No idempotency or destructiveness markers.
Get details about a specific assignment
Get information about a specific course
Get current user information
Get details about a specific message
Get all grades for the current user across all sections
Get information about a specific section
Get grades for a specific section
Get information about a specific user
8 list_* and get_* tools have no documented input schemas despite accepting parameters. get_me, list_my_sections, get_my_grades, list_assignments, list_inbox, list_sent_messages, list_my_events, list_discussions lack visible parameter definitions.
All tool descriptions are 10-50 characters and lack context for LLM selection. Descriptions do not explain WHEN to use each tool vs similar ones (e.g., get_user vs get_me, list_assignments vs list_discussions). No guidance on prerequisites or dependencies.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 44 | 2026-07-28+ | v2 |
List all assignments in a section
List all discussions in a section
List messages in the user's inbox
List all events for the current user
List all sections the current user is enrolled in
List messages sent by the current user
No output schemas documented. LLMs cannot plan downstream tool calls or extract required fields (e.g., section_id from list_my_sections to pass to list_assignments). Baseline: 100% of A+ tools document return types.
No pagination support documented for list_* tools (list_my_sections, list_assignments, list_inbox, list_sent_messages, list_my_events, list_discussions). No limit, offset, or cursor parameters visible. Large result sets will exhaust context window.
No error handling guidance. No recovery hints (e.g., 'User not found. Try search_users() with a partial name.'). No categorization of errors as retryable, user-fixable, or fatal. Agents cannot determine next steps on failure.