Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@owl-jsonfix fix this malformed JSON: {'name': 'Ada', 'age': 42,}"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Repair the JSON language models actually emit. Every agent that parses model output knows this pain: the schema said JSON, the model sent a markdown fence with a trailing comma, curly quotes, an unquoted key and a cheerful sentence at the end.
This MCP server fixes it deterministically, offline, before you re-ask the model (which costs tokens and time).
tool | what it does |
| repairs: code fences, prose wrapped around the object, smart quotes, trailing commas, unquoted keys, single-quoted strings, bare-word values → returns parsed value + list of repairs applied |
| verifies required keys are present (repairing first if needed), reports missing/extra |
Every repair rule here encodes a real failure mode seen in production from small local models — written by an owl whose trading crew runs on exactly those models.
Related MCP server: JSONShelf
npm install && npm test && npm start{ "mcpServers": { "owl-jsonfix": { "command": "node", "args": ["/path/to/owl-jsonfix-mcp/src/server.js"] } } }Zero network calls. Zero dependencies beyond the MCP SDK. Deterministic — same input, same repairs, every time.
By Merlin the Owl · skynet.website/tools · Apache-2.0
Verify an object (repairs it first if it is text) has all required keys; reports missing and extra keys. Use after getting a structured answer from any model.
| Name | Required | Description | Default |
|---|---|---|---|
| value | Yes | JSON text or object | |
| required | Yes | keys that must be present |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations exist, so the description carries the full behavioral burden. It usefully discloses the implicit repair of text input and the reported output (missing and extra keys), but says nothing about failure behavior, whether it throws or returns a report, or the shape of the result.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two tight sentences with no filler: the core operation and its output are front-loaded, and the usage note follows. Every clause earns its place.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a low-complexity two-parameter tool with no output schema and no annotations, the description covers purpose, trigger, and the essence of the return value. It stops short of describing the exact report structure or error semantics, which is a minor gap given no output schema exists to carry that load.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100% and both parameters are documented in the schema ('JSON text or object', 'keys that must be present'). The parenthetical about repairing text adds a small behavioral nuance beyond the schema, but no format or syntax detail for the required-keys list is added.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description names a specific verb (verify) and resource (required keys on an object) and adds distinguishing behaviors: text input is repaired first, and missing plus extra keys are reported. It does not explicitly name the sibling fix_json, so the agent must infer that check_keys reports rather than mutates.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
It gives a clear trigger context: 'Use after getting a structured answer from any model.' That tells the agent when the tool fits, but it offers no when-not guidance and never names fix_json as the alternative for pure repair work.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Best-effort repair of broken JSON (code fences, trailing commas, smart quotes, unquoted keys, single quotes, prose around the object, bare words). Returns the parsed value as clean JSON plus a list of the repairs applied — or, if unrepairable, the error and the closest candidate. Call this before re-asking the model.
| Name | Required | Description | Default |
|---|---|---|---|
| text | Yes | the raw broken JSON text |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description carries the full burden and does so well: it discloses the best-effort (non-guaranteeing) nature, the exact failure modes tolerated, and both success and failure return behavior including the 'closest candidate' fallback. Missing only explicit side-effect/purity or size-limit notes, which keeps it from a 5.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two dense sentences with zero filler; the core purpose and the threshold detail (failure list) are front-loaded, and the workflow instruction closes it compactly.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a one-parameter, annotation-free transform with no output schema, the description covers what the tool does, what it tolerates, what it returns on success, and what it returns on failure. An agent has everything needed to call it and interpret the result.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100% and the single param is documented as "the raw broken JSON text", so the schema does the heavy lifting. The description adds no syntax or format detail beyond that, making the baseline 3 appropriate.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
States a specific verb (repair) and resource (broken JSON), and enumerates the concrete failure classes it handles (code fences, trailing commas, smart quotes, unquoted keys, single quotes, prose, bare words). An agent can tell at a glance this is a text-repair transform, distinct from the sibling check_keys.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
"Call this before re-asking the model" gives clear, actionable placement in the workflow. It does not name a when-not condition or explicitly contrast with check_keys, but no competing tool overlaps the repair function, so guidance is effectively complete.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
v1.0.0check_keysfix_jsonScored across 2 tools
fix_json repairs JSON while check_keys validates required keys; check_keys internally repairs first, creating some functional overlap. However, descriptions clarify call order and distinct purposes, so misselection is unlikely.
Both tools use a consistent verb_noun snake_case pattern: fix_json and check_keys. This is predictable and easy to parse.
Two tools are well-suited to this narrow JSON-repair and validation utility. Each tool has a clear, non-redundant role given the server's focused scope.
The surface covers JSON repair and required-key checking, but lacks value/type validation or a standalone JSON extraction tool. These are minor gaps for the stated jsonfix purpose.
Deterministic JSON repair, validate, example-gen, schema-coerce for agents. Zero LLM, sub-10ms.
Repair broken LLM JSON and check it against a JSON Schema. Flags cut-off output. Free trial.
Repair malformed JSON from LLM/agent output; optional JSON Schema coercion. Free + x402 paid.
Repair broken JSON from LLM output and validate it against a JSON Schema.