Loading ToolBench
Free, open-source AI resume tailor — automatically rewrite your resume to match any job description for ATS optimization
Single tool 'tailor_resume' has a reasonable description (156 chars) and input schema with three parameters. However, critical gaps severely limit production readiness: (1) NO output schema documented, LLMs cannot plan downstream actions or extract results; (2) parameter descriptions are minimal (10-30 chars) and lack format/constraint guidance; (3) no error handling guidance, failures return raw error messages with no recovery path; (4) tool is stateful (depends on OpenAI API key in env) but no idempotency guarantees; (5) no pagination, rate limiting, or result size caps despite potentially large resume JSON inputs. The implementation is a Next.js API route, not an MCP server, no evidence of MCP protocol compliance in the codebase.
Rewrite and enrich resume sections to highlight achievements, skills, and responsibilities relevant to a job description. Preserves factual data while modifying summary, experience bullets, and skills for ATS optimization.
No output schema documented. LLMs cannot infer what fields the tailored resume contains or plan downstream tool calls. Critical for agent composition.
Parameter descriptions are too brief (10-30 chars) and lack format/constraint guidance. 'Resume data in JSON format' does not specify required keys, nesting depth, or validation rules. LLMs will pass malformed input.
No error handling guidance. Failures return raw error messages ('Failed to parse GPT response as JSON') with no recovery path. LLMs cannot self-correct or suggest alternatives.
Codebase is a Next.js web app (package.json shows 'next', 'react', 'react-dom'), not an MCP server. No MCP protocol implementation visible. Transport, capabilities, and tool registration are missing.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 36 | 2026-07-28+ | v2 |
OpenAI API key exposed as environment variable with no secret injection pattern. If this were an MCP server, credentials would leak into agent traces and logs.