Regex testing, explanation, and generation REST API + MCP server
Regex-toolkit has 8 well-defined tools with complete input schemas and consistent naming (verb_noun pattern). All tools have descriptions (avg 85 chars, within baseline 34-392). However, descriptions are generic and lack actionable context for LLM selection. Parameter descriptions are minimal (e.g., 'Regex flags' without format guidance). No output schemas documented. Error handling returns raw error text without recovery guidance. Missing tool annotations (readOnlyHint/destructiveHint). One tool (regex_replace) is destructive but unmarked.
Benchmark regex execution speed.
Escape special regex characters in a string.
Explain a regex pattern in plain English, token by token.
List all common regex patterns, or get a specific one by name.
Replace matches of a regex in text.
Split text using a regex pattern.
Test a regex against text. Returns all matches, groups, and indices.
No output schemas documented. LLMs cannot predict response structure or plan downstream calls. E.g., regex_test returns matches/groups/indices but this is not formally specified.
Error handling returns raw error text without recovery guidance. E.g., 'Pattern "xyz" not found' does not suggest calling regex_patterns to discover available patterns.
regex_replace is destructive (modifies text) but has no tool annotation (destructiveHint) and description does not warn of side effects. Agents cannot distinguish safe vs. unsafe tools.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 55 | 2026-07-28+ | v2 |
Check if a regex pattern is syntactically valid.
Parameter descriptions lack format guidance. E.g., 'Regex flags' does not explain valid values (g, i, m, s, u, y) or that they are case-sensitive. LLMs may pass invalid flags.
Tool descriptions are generic and do not explain WHEN to use each tool vs. alternatives. E.g., regex_explain vs. regex_validate both analyze patterns but serve different purposes, not distinguished in descriptions.