Customizable AI Assistant Discord Bot with MCP tool integration, multi-provider LLM support, and advanced persona management
TomoriBot is not an MCP server, it is a Discord bot with a custom tool system. Tool definitions exist only as string constants in toolNotices.ts with no visible input schemas, parameter descriptions, or output documentation. No tool registration mechanism, no JSON Schema, no error handling patterns. The 13 tools are listed by name and brief description only, with no structured metadata. This fails the fundamental requirement that MCP tools must have complete, machine-readable schemas.
Read and extract text from documents
Process and extract keyframes from GIFs
Analyze images using vision models
Edit images using inpainting or outpainting
Generate images from text prompts
Search for images
Call tools from MCP servers
No input schemas visible. All 13 tools lack JSON Schema definitions with parameter types, constraints, and descriptions.
Descriptions are trivial (10-30 chars). All tool descriptions are single short sentences with no context on WHEN to use, WHAT parameters are needed, or WHAT the output contains.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 32 | 2026-07-28+ | v2 |
Search for news articles
Generate videos from text prompts
Search for videos
Fetch and read webpage content
Search the web for information
Download and process YouTube videos
No parameter documentation. Tool definitions do not list parameters, their types, constraints, or descriptions. LLMs cannot infer what inputs each tool accepts.
No output schema documentation. Tools do not declare what fields they return, their types, or structure. Agents cannot plan downstream calls or extract data reliably.
No error handling guidance. Tools do not document error conditions, recovery steps, or whether errors are retryable. Agents have no way to respond to failures.