Loading ToolBench
MCP server that executes web application automation tasks using a self-driving web application action driver client
This server has critical definition gaps. The single tool 'execute_skill_action' has a minimal description (47 chars, below the 10-1024 baseline), lacks output schema documentation, and the input schema is present but sparse. The mcp_server.py file shows only a stub that prints JSON and does not implement actual MCP protocol message handling. Tool definitions appear inferred from client.py rather than explicitly registered via proper MCP mechanisms. No error handling, no parameter validation guidance, no security considerations documented.
Execute a web application automation skill action
Tool description is too short (47 chars) and lacks context. Does not explain WHEN to use this tool, what prerequisites exist, or what the agent should expect in return. LLMs cannot determine tool selection without richer guidance.
Output schema is not documented. The tool returns a dict with 'actions_executed_count', 'goal_accomplished', and 'final_state_summary', but this structure is not declared in the tool definition. LLMs cannot plan downstream actions without knowing what fields to expect.
Input parameters lack descriptions. 'target_url' and 'task_objective' have type and name but no guidance on format, constraints, or examples. Parameter descriptions must explain what values are valid and what the parameter controls.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 30 | 2026-07-28+ | v2 |
No error handling or recovery guidance. The tool provides no documentation of failure modes, retryability, or what the agent should do if the web automation fails. Error responses must guide the LLM on next steps.
Tool definition is inferred from client.py, not explicitly registered via MCP protocol. The mcp_server.py stub does not implement proper tool registration, message handling, or protocol compliance.