Loading ToolBench
Audits GitHub repos and profiles for hiring signal — as an MCP server (audit_repo, recruiter_scan tools) and as a standalone CLI.
Two tools with clear, domain-specific descriptions (194 - 250 chars, within baseline 34 - 392 range). Input schemas present with Zod validation and parameter descriptions. However, output schemas are undocumented, responses are JSON-stringified text blobs with no declared structure. Error handling is minimal; no recovery guidance or actionable error messages. Tool names are verb-noun but lack clarity on what 'audit' and 'scan' return. No pagination, no per-item error reporting, no idempotency hints.
Gathers hiring-signal facts about a GitHub repository (README presence, CI, license, topics, commit-message quality, activity recency) and a rule-based verdict. Use this data to reason about portfolio-readiness — this tool reports facts, it does not itself judge code quality or architecture.
Gathers the checkable signals a technical recruiter's 60-90 second scan would notice about a GitHub profile (bio presence, pinned-repo count, activity recency, public repo count) and a rule-based pass/maybe/reject lean. Use this data alongside your own read of the actual repos for the full picture.
Output schemas undocumented. Tools return JSON-stringified text with no declared field structure. LLMs cannot plan downstream calls or extract specific fields reliably.
No error handling or recovery guidance. If GitHub API fails or rate-limits, tools return raw errors with no actionable next steps for the LLM.
Tool names 'audit_repo' and 'recruiter_scan' are vague about return structure. 'audit' could mean many things; 'scan' is generic. Names like 'analyze_repo_hiring_signals' or 'evaluate_profile_readiness' would better convey intent.
No pagination or result limits documented. If a repo has thousands of commits or a user has hundreds of repos, responses could balloon. No limit parameter or cursor guidance.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 51 | 2026-07-28+ | v2 |
No tool annotations (readOnlyHint, idempotentHint). Both tools are read-only and idempotent, but this is not declared in the schema, forcing LLMs to infer safety.