- What it does
- Run a regular expression (Python flavor) against sample strings with a time cap and get every match, span and group, or the pattern error, with a signed record.
- Who it is for
- Agents and their builders.
- Start in one step
- Ask your agent to call
regex_testat llpro.ai/mcp (free, no sign-up). This page explains the tool; it has no form to run it. - What to expect
- Testing: open to try. Free
New to this? Start here.
What it does
Regular Expression Test is for agents and developers that write or review regular expressions for validation, parsing or log search and need to see exactly what they match before shipping them. Run a regular expression against sample strings, with a time cap. Send pattern, samples (up to 50 strings) and optionally mode (search, fullmatch or all), and the flags ignore_case, multiline, dotall, ascii. Get back, for each sample, whether it matched and every match with its start and end (counted in characters), the matched text and each capture group (number, name, span, text), or a clear error if the pattern is not valid. The flavor is Python (the 'regex' module in VERSION0 mode, a superset of Python re); the answer says if plain Python re would also accept the pattern. A pattern that backtracks for more than 0.5 s on a sample is reported as timed_out, not run forever. It is one of the developer department tools LLPRO builds for the work agents do across business software (GitHub, Vscode, Postman). Every answer carries the written rule it used, what it does not do, its sources with dates, and an Ed25519 signature you can check against the key below.
- Who it is for
- Agents and developers that write or review regular expressions for validation, parsing or log search and need to see exactly what they match before shipping them.
- Price
- Free, no sign-up: 120 calls an hour per address and 3,000 an hour across all callers. Paid calls are planned and not charging.
- Limits
- Up to 30,000 bytes per call, samples up to 50 entries; larger requests are refused, never cut. Unknown fields are refused with the list of accepted ones. Nothing you send is stored by the tool; no model and no network are used. Does not test other flavors: JavaScript, PCRE, Java, .NET, Go (RE2) and POSIX differ in $ and \Z, named-group syntax ((?P<n>) here, (?<n>) in JavaScript), \d and \w on non-ASCII text, lookbehind limits and which escapes exist. A pattern that works here may fail there. Does not explain or rewrite the pattern and does not judge whether it is a good validator (an email or phone regex is not proof an address exists). Replacements and splits are not offered; only matching. Time cap results depend on the speed of the server: a pattern that is slow but under the cap is answered, one over 0.5 s on a sample is timed_out, and a borderline one may land either side. A pattern that needs over 1 GB to compile or is otherwise too heavy is reported as pattern_error. The regex module's fuzzy matching ({e<=1}) and (?V1) are not blocked, only capped by the time and memory limits. Limits: pattern 500 characters, 50 samples of up to 2000 characters, 100 matches per sample and 500 in all, whole request 30,000 bytes. No personal data: samples are screened for emails, web addresses with a scheme, Social Security and phone shapes and card numbers (digits passing the Luhn check), and the pattern for Social Security, phone and card numbers; use made-up values. Nothing is stored; matched text is echoed back as data.
For developers: try it
POST https://llpro.ai/mcp
{"jsonrpc": "2.0", "id": 1, "method": "tools/call", "params": {"name": "regex_test", "arguments": {"pattern": "(\\d{4})-(\\d{2})-(\\d{2})", "samples": ["Due 2026-11-22 or 2026-12-01", "no date"], "mode": "all"}}}For agents
Tool regex_test on https://llpro.ai/mcp and in the tools list. Signed answers are checked with the public key below (signer llpro1:n53oivccov6migyvs7pwsovtbi): an Ed25519 signature over the observation as canonical JSON (sorted keys, no spaces).
Public key (base64): a0LEEWyhyl/2bSr4Mkb7b4SKB+I815t8sc4x6MGfsCQ=
For agents that search MCP registries for regex, regular expression tester, regex match groups tools (listed in MCP registries, served by the LLPRO gateway: MCP, agent-to-agent and REST). One fixed, written rule per call, a signed record, no account needed.
Next
- Connect your agent: Use this and the other LLPRO tools from your own agent.
- Get your agent: No agent yet? Get one that asks first.
- Hand the whole job to LLPRO: Testing a pattern is one step of writing a parser or validator. Tell LLPRO the whole job; it routes you to the right signed tools (JSONPath query, JSON Schema validate).
For agents: this page as Markdown · JSON · llms.txt · MCP endpoint llpro.ai/mcp · agent card