Loading ToolBench

Reading the current corpus…

genpark-visual-screen-blackboard-explainer-skill MCP server · F · ToolBench

genpark-visual-screen-blackboard-explainer-skill

github.com/alphaparkinc/genpark-visual-screen-blackboard-explainer-skill

MCP server that explains visual screens and system architecture diagrams by analyzing screenshots and generating annotated diagrams with identified UI elements

2026-07-28+STDIO1 tools

16/100

Definition Quality15

Protocol Readiness5

Supportability25

Spec posture

Inferred effective spec 2026-07-28+·current baseline 2026-07-28

Static source inference · medium confidence

Current-spec patterns detected

Summary

Critical failures across all dimensions. The single tool 'execute_skill_action' has no visible input schema, no parameter descriptions, and a vague 1-line description (18 chars). The mcp_server.py file does not implement the MCP protocol, it merely prints a JSON status message and exits. No tool registration, no schema validation, no error handling. This is not a functional MCP server.

Tools (1)

execute_skill_actionread onlysource verified15/100

Execute a skill action (declared in MCP server initialization)

Issues (5)

CRITICALtool ↗

No input schema visible for execute_skill_action. Tool definition is inferred from a single print statement, not from explicit MCP tool registration.

execute_skill_action

CRITICALtool-description ↗

Tool description is 18 characters ('Execute a skill action (declared in MCP server initialization)'), below the 20-char minimum. Provides no context for when/why to use the tool or what it returns.

execute_skill_action

CRITICALtool-description ↗

No parameter descriptions. The tool accepts parameters (screenshot_uri, user_query inferred from client.py) but the MCP server does not declare them with types or descriptions.

execute_skill_action

CRITICALtool ↗

mcp_server.py does not implement the MCP protocol. It prints a status message and exits instead of entering a request/response loop. No message handling, no tool invocation, no error reporting.

execute_skill_action

Recommendations

Implement a proper MCP server message loop (JSON-RPC 2.0 over STDIO) that listens for initialize, list_tools, call_tool, and other MCP requests.
Register execute_skill_action with a complete JSON Schema input definition: {type: 'object', properties: {screenshot_uri: {type: 'string', description: '...'}, user_query: {type: 'string', description: '...'}}, required: ['screenshot_uri']}.
Expand tool description to 50 - 200 characters explaining what the tool does, when to use it, and what it returns. E.g., 'Analyzes a screenshot and explains the system architecture, UI elements, and potential bottlenecks shown. Returns annotated diagrams and identified components.'
Document the output schema in the tool definition: {type: 'object', properties: {annotated_diagram_md: {type: 'string'}, identified_ui_elements: {type: 'array'}, explanation_summary: {type: 'string'}}}.
Add error handling: if screenshot_uri is invalid or the analysis fails, return a structured error with recovery guidance (e.g., 'Invalid image format. Supported: PNG, JPG, GIF').

Score history

Overall score trend

First recorded score · v2 rubric

16/100

Scored	Grade	Overall	Spec posture	Rubric
2026-09-23	F	16	2026-07-28+	v2