jsonpad MCP server
Which servers in the official MCP registry are jsonpad's own, whether they answer, and what they offer. A program checked each of these, nobody copied them from a product page.
read 2026-10-15
- Registry nameio.jsonpad/docsverified in the MCP registry
The registry name sits under jsonpad.io. The registry only lets a publisher use that namespace after proving control of the domain through DNS or HTTP.
- Stateanswerschecked 2026-10-15
- Tools14
- Server calls itselfjsonpad-docs
- Server version0.1.0
- Protocol version2025-06-18
On 2026-10-15 the server answered initialize and tools/list without a sign-in. First answer in 76 ms.
- Whether its tools do the product's job has not been judged yet.
- Checked on 3 of the 7 days it takes to judge reliability.
Said by rules from the measurements above. The likely first call is a model's pick from the tool list; read its description before relying on it.
- INITE MCP scoreafter a weeka score needs 7 daily checks: 3 so far
- Answeredmeasuring3 of the 7 daily checks it takes to print a share
- Median first answer246 ms
- Space in an agent's context≈ 2,981 tokensthe definitions of all its tools (14), which an agent carries on every turn
- Tool definitionsA · 100 of 100
- Typed results14 of 14 toolsdeclared outputSchema: results come typed, read once
- Every tool passes all six checks.
Worth a look
None of the five warning signs in these definitions.
A program checks the server once a day: initialize and tools/list, no model. Tokens are the definitions' length at about four characters a token. The grade averages six checks of every tool's name, description, schema and annotations; it says how well the tools are described, not whether they work. The score weighs what was measured and nothing else: answering (30%), doing the product's job (25%), definitions (20%), context (10%) and use (15%). A part that was not measured is left out, and a hidden instruction caps the score at 40. The state of vendors' MCP servers: method and numbers →
search_docsSearch the JSONPad documentation, the JavaScript SDK references and the command line tool reference. Returns the best-matching sections with a snippet and their jsonpad.io URL; read one with read_doc. Exact identifiers work well: error…
read_docRead a documentation page as markdown, or one section of it. Give a page slug ("indexing"), a /docs path, a jsonpad.io URL (with or without .md, and with an optional #anchor), "sdk/jsonpad-sdk", "sdk/jsonpad-realtime-sdk", "cli/reference"…
list_docsList the documentation pages by section, with a description of each: the table of contents.
list_api_endpointsList the REST API endpoints an API token can use, as method, path and title. Get one in full with get_api_endpoint.
get_api_endpointThe full contract of one REST API endpoint: parameters, headers, request body, responses and examples, and the JavaScript SDK method that calls it. Find it by id (the docs page slug, e.g. "item-restore"), by method and path (templates or…
get_sdk_methodThe signature, description and example of a method in the JavaScript SDK (@basementuniverse/jsonpad-sdk) or the realtime SDK, and the REST endpoint it calls. Find it by name ("restoreItem") or by what it does.
get_cli_commandThe usage, arguments and options of a jsonpad command line tool (@basementuniverse/jsonpad-cli) command, and whether it works offline. Find it by name ("rules test", "jsonpad sync-schema --dry-run") or by what it does.
lookup_errorWhat a JSONPad API error code means, its HTTP status, and the guides that explain it. Give the numeric code (10013) or the name (QUOTA_EXCEEDED), as found in an error response's "code" and "name".
get_plan_limitsThe limits of each public plan (Free, Indie, Pro, Scale): requests per month and minute, the minimum gap between requests, storage, item size, versions kept, tokens, identities, realtime connections, webhooks and flows. null means…
check_write_rulesCompile a list's write rules with the same engine the API uses, and run their tests (a rules-tests-v1 document). Returns every diagnostic with its line and column, and for each failing test what it expected, what happened, and how each…
eval_write_ruleCheck one write against a rule set, as the API would, and say whether it would be allowed, denied (403) or failed (400), which statement decided it, and (with trace) what every expression evaluated to. The write is shaped like a rule test…
check_flowCompile a flow document (flows-v1) with the same engine the API uses, and run its tests (flow-tests-v1) against in-memory data. Returns diagnostics (with JSON pointers into the document), what the flow reads and writes, and why each…
validate_sync_documentCheck a schema sync document (sync-v1) as far as possible without an account: against the JSON schema the API uses, the API's rules for keys and indexes, and every rule set and flow in it with their tests. Documents that reference files…
validate_token_permissionsCheck an API token's permission rules against the schema the API uses, say which rule shape a broken one was meant to be, and warn about valid rules that probably don't do what was meant (rule order, restore without view, sync-schema…
How to connect
Add the server to your MCP client config. Claude Desktop, Cursor and most other clients read this shape:
{
"mcpServers": {
"docs": {
"type": "http",
"url": "https://mcp.jsonpad.io/docs"
}
}
}Add it
Claude Code, in a terminal
claude mcp add --transport http jsonpad https://mcp.jsonpad.io/docsclaude.ai, ChatGPT and other clients: add https://mcp.jsonpad.io/docs as a custom connector in the settings.
Also shipped through: npm
- io.jsonpad/apianswers, requires sign-inverified in the MCP registry
Each line is something a program found when it went looking: the registry entry, the server's own answer, a spec that parses. A missing line means we did not find it, not that it does not exist.
Cookies
Before we count this visit
Analytics and ad measurement tell us which pages and which of our ads do their job: Google Analytics and Google Ads, the Meta pixel, Yandex Metrica, and one first-party id of our own. Nothing is stored until you answer, and declining changes nothing about the site.