Loading ToolBench
Terminal-based AI assistant for software development with LSP integration, code analysis, and MCP tool support
VideoCoder exhibits severe definition quality gaps across all dimensions. Tool descriptions are present but lack LLM-optimization guidance (e.g., agent tool description is 1000+ chars, violates 10-1024 char baseline). No visible input schemas in source code for any of the 12 tools, only tool names and descriptions are evident. Parameter types and constraints are completely absent. Error handling guidance is missing. The 'agent' tool description is excessively verbose and includes implementation details rather than user-facing intent. No output schemas documented. Security-critical tools (bash, edit, write, patch) lack permission gates or destructive operation warnings despite high-risk profiles.
Launch a new agent that has access to the following tools: GlobTool, GrepTool, LS, View. When you are searching for a keyword or file and are not confident that you will find the right match on the first try, use the Agent tool to perform the search for you. For example: - If you are searching for a keyword like "config" or "logger", or for questions like "which file does X?", the Agent tool is strongly recommended - If you want to read a specific file path, use the View or GlobTool tool instead of the Agent tool, to find the match more quickly - If you are searching for a specific class definition like "class Foo", use the GlobTool tool instead, to find the match more quickly Usage notes: 1. Launch multiple agents concurrently whenever possible, to maximize performance; to do that, use a single message with multiple tool uses 2. When the agent is done, it will return a single message back to you. The result returned by the agent is not visible to the user. To show the user the result, you should send a text message back to the user with a concise summary of the result. 3. Each agent invocation is stateless. You will not be able to send additional messages to the agent, nor will the agent be able to communicate with you outside of its final report. Therefore, your prompt should contain a highly detailed task description for the agent to perform autonomously and you should specify exactly what information the agent should return back to you in its final and only message to you. 4. The agent's outputs should generally be trusted 5. IMPORTANT: The agent can not use Bash, Replace, Edit, so can not modify files. If you want to use these tools, use them directly instead of going through the agent.
Execute bash commands
Get diagnostics from LSP servers
Edit files with LSP support
No input schemas visible in source code for any tool. Only tool names and minimal descriptions provided. Parameter types, constraints, and validation rules completely absent.
Destructive tools (bash, edit, write, patch) lack security annotations, permission gates, or confirmation patterns. No dry-run support or destructive operation warnings.
Agent tool description exceeds 1000 characters (baseline max 1024), contains implementation details and usage notes that belong in documentation, not tool descriptions. Violates LLM-optimization principle.
Inferred effective spec: 2026-07-28+.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 31 | 2026-07-28+ | v2 |
Fetch remote content
Search for files using glob patterns
Search for text patterns in files
List directory contents
Apply patches to files with LSP support
Search code using Sourcegraph
View file contents with LSP support
Write content to files with LSP support
No output schemas documented for any tool. LLMs cannot plan downstream tool calls or extract required fields without knowing response structure.
Tool descriptions are generic one-liners (e.g., 'Execute bash commands', 'Edit files with LSP support'). No guidance on WHEN to use each tool, prerequisites, or error recovery paths.