Loading ToolBench
MCP server that gives Claude Code your team's engineering standards — git, Rails, testing, deployment, code review, frontend
Server has 4 tools with clear verb-noun naming (get_standard, review_branch_name, review_commit_message, staff_engineer_review). All tools have descriptions and input schemas with enums/types. However, descriptions are generic and lack actionable guidance for LLM selection. Parameter descriptions are minimal (e.g., 'The branch name to review' lacks format/constraint details). Output schemas are not documented, responses are free-text rather than structured objects. Error handling is absent; tools return text feedback but no recovery guidance. No tool annotations (readOnlyHint, etc.). Composition is reasonable but tools lack chaining IDs and pagination support.
Retrieve a specific engineering standard. Available standards: git, code_review, pr, staff_engineer, testing, rails, frontend, deployment, incident_response, observability, api_design, database_design, architecture_decisions, technical_debt, standup
Check if a git branch name follows conventions (issue ID prefix, kebab-case)
Check if a commit message follows conventions (imperative mood, subject length, structure)
Run through the staff engineer thinking checklist for a given context (pre-code, pre-pr, post-merge, incident)
Output schemas not documented. Tools return free-text responses (text content blocks) without structured field definitions. LLMs cannot plan downstream calls or extract typed data.
Parameter descriptions lack actionable constraints. 'The branch name to review' does not specify format, length, or valid characters. 'Brief description' is vague. LLMs cannot validate inputs before calling.
No error handling or recovery guidance. Tools return text feedback (issues/suggestions) but do not categorize errors as retryable, user-fixable, or fatal. No guidance on next steps.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 52 | 2026-07-28+ | v2 |
Tool descriptions are generic and do not explain WHEN to use each tool vs. alternatives. 'Check if a git branch name follows conventions' does not clarify when to call this vs. get_standard for git conventions.
No tool annotations (readOnlyHint, idempotentHint). All tools are read-only and idempotent, but this is not declared in the schema. LLMs cannot infer safety properties.