- What it does
- Run a JSONPath (RFC 9535) query on a JSON document and get the matching values with their normalized paths, or the exact syntax error, with a signed record.
- Who it is for
- Agents and their builders.
- Start in one step
- Ask your agent to call
jsonpath_queryat 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
JSONPath Query is for agents and developers that pull fields out of API responses, config files or tool output with JSONPath and want the standard's answer, not one library's dialect. Run a JSONPath query (RFC 9535) on a JSON document you send. Send query, for example $.store.book[?@.price < 10].title, and document, the JSON text. Get back how many nodes matched and, for each, its value and its normalized path (for example $['store']['book'][0]['title']), in the order RFC 9535 defines. Supports name, index, wildcard, slice and filter selectors, the descendant segment .., and the standard functions length, count, match, search and value. A query that is not valid RFC 9535 gives verdict query_error with the reason and the position. Limits: query 1,000 and document 15,000 characters. It is one of the developer department tools LLPRO builds for the work agents do across business software (GitHub, Postman, Aws). 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 pull fields out of API responses, config files or tool output with JSONPath and want the standard's answer, not one library's dialect.
- 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; 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 accept the older Goessner JSONPath dialect or other libraries' extensions: a script expression ?(@.price<10), a bare = , 'in' operators, @.length, $..['a','b'] sets with filters inside without ?, and other non-RFC syntax are query_errors. Does not change the document (no update or delete) and does not fetch anything. Numbers are doubles: a match prints a value like 1.0 as 1.0, and 1e2 as 100.0. Object members are visited in the order they appear in your text, which RFC 9535 leaves unspecified; tell your reader if order matters. I-Regexp patterns in match and search are run with a 0.2 s limit per test (counted in the 2.5 s of the call; a test that runs out ends the query as limit_reached) and a size cap (pattern up to 200 characters; a pattern with repeat counts that would expand beyond 10,000 is treated as not valid, so false). Limits: query 1,000 and document 15,000 characters, 32 levels of nesting in the document and 8 in a query, 200,000 steps or 2.5 s of work per call, the first 100 matches listed (the count is complete, up to the work limit) and about 60,000 characters of values. No personal data: document strings are screened for emails, Social Security and phone shapes and card numbers, which are refused; the query is screened for Social Security, phone and card numbers; use made-up values. Values are echoed back as data.
For developers: try it
POST https://llpro.ai/mcp
{"jsonrpc": "2.0", "id": 1, "method": "tools/call", "params": {"name": "jsonpath_query", "arguments": {"query": "$.store.book[?@.price < 10].title", "document": "{\"store\": {\"book\": [{\"category\": \"reference\", \"title\": \"Sayings\", \"price\": 8.95}, {\"category\": \"fiction\", \"title\": \"Sword\", \"price\": 12.99}, {\"category\": \"fiction\", \"title\": \"Moby\", \"isbn\": \"0-553\", \"price\": 8.99}], \"bicycle\": {\"color\": \"red\", \"price\": 399}}}"}}}For agents
Tool jsonpath_query on https://llpro.ai/mcp and in the tools list. Signed answers are checked with the public key below (signer llpro1:cih53gnnnibqhwkcqctu5fbbc4): an Ed25519 signature over the observation as canonical JSON (sorted keys, no spaces).
Public key (base64): i6Z3pQPvsAzLA92Jl1+bQ0YyOryMLCh/nIUfL46X4EQ=
For agents that search MCP registries for jsonpath, RFC 9535, json query 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: Querying JSON is one step of reading an API or a file. Tell LLPRO the whole job; it routes you to the right signed tools (JSON Schema validate, regex test).
For agents: this page as Markdown · JSON · llms.txt · MCP endpoint llpro.ai/mcp · agent card