Loading ToolBench
Simple MCP server for Claude Code one-shot execution
Single tool 'claude_code' has a verbose description (500+ chars) but lacks critical structure. Input schema is present with two parameters (prompt, workFolder), but workFolder is optional despite being marked 'Mandatory' in description, a dangerous contradiction. No output schema documented. Parameter descriptions are minimal. The tool bundles file ops, Git, terminal, web search, and GitHub into one massive responsibility, violating single-concern principle. Risk marked IRREVERSIBLE but no confirmation/dry-run pattern implemented. No error recovery guidance provided.
Claude Code Agent: Your versatile multi-modal assistant for code, file, Git, and terminal operations via Claude CLI. Use `workFolder` for contextual execution. • File ops: Create, read, (fuzzy) edit, move, copy, delete, list files, analyze/ocr images, file content analysis └─ e.g., "Create /tmp/log.txt with 'system boot'", "Edit main.py to replace 'debug_mode = True' with 'debug_mode = False'", "List files in /src", "Move a specific section somewhere else" • Code: Generate / analyse / refactor / fix └─ e.g. "Generate Python to parse CSV→JSON", "Find bugs in my_script.py" • Git: Stage ▸ commit ▸ push ▸ tag (any workflow) └─ "Commit '/workspace/src/main.java' with 'feat: user auth' to develop." • Terminal: Run any CLI cmd or open URLs └─ "npm run build", "Open https://developer.mozilla.org" • Web search + summarise content on-the-fly • Multi-step workflows (Version bumps, changelog updates, release tagging, etc.) • GitHub integration Create PRs, check CI status • Confused or stuck on an issue? Ask Claude Code for a second opinion, it might surprise you! **Prompt tips** 1. Be concise, explicit & step-by-step for complex tasks. No need for niceties, this is a tool to get things done. 2. For multi-line text, write it to a temporary file in the project root, use that file, then delete it. 3. If you get a timeout, split the task into smaller steps. 4. **Seeking a second opinion/analysis**: If you're stuck or want advice, you can ask `claude_code` to analyze a problem and suggest solutions. Clearly state in your prompt that you are looking for analysis only and no actual file modifications should be made. 5. If workFolder is set to the project path, there is no need to repeat that path in the prompt and you can use relative paths for files. 6. Claude Code is really good at complex multi-step file operations and refactorings and faster than your native edit features. 7. Combine file operations, README updates, and Git commands in a sequence. 8. Claude can do much more, just ask it!
Tool bundles 8+ distinct responsibilities (file ops, code generation, Git, terminal, web search, GitHub, multi-step workflows, analysis). Violates single-concern principle. Should split into separate tools: create_file, edit_file, run_git_command, run_terminal_command, search_web, create_github_pr, etc.
Parameter 'workFolder' marked 'Mandatory' in description but defined as optional (not in 'required' array). This contradiction will cause LLMs to omit it, leading to execution failures. Either make it required in schema or remove 'Mandatory' from description.
No output schema documented. LLMs cannot predict what fields to expect from claude_code execution (success/failure, stdout, stderr, exit code, etc.). Impossible to chain downstream tool calls or extract structured data.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 43 | 2026-07-28+ | v2 |
Tool marked IRREVERSIBLE (delete, Git push, terminal commands) but no dry-run, confirmation, or multi-round-trip (MRTR) pattern implemented. Agents cannot preview changes before execution. No recovery guidance if destructive operation fails mid-sequence.
Parameter descriptions are minimal. 'prompt' lacks guidance on format, length, or examples. 'workFolder' says 'absolute path' but does not explain consequences of wrong path, whether it must exist, or how relative paths are handled. LLMs will guess.