JSON
A lightweight data-interchange format that uses curly braces to wrap key-value pairs and square brackets to denote arrays. Its clear structure makes it easy for both machines to parse and humans to read, and it has become the de facto standard for Web API communication.
Core Facts
The core mechanism of Function Calling is that developers describe external function signatures in JSON Schema format, and the model outputs structured JSON invocation instructions that conform to the schema specification.
The basic principle of tool calling is that the model identifies scenarios requiring external capabilities, outputs a structured function call request (usually in JSON format), and after the external system executes it, the result is returned to the model to continue reasoning.
Function Call does not actually execute code; the model only outputs a structured JSON string, and the actual execution is performed by the host program.
Timeline (last 90 days)
In a workflow platform, the DSL typically defines the flow structure in JSON or YAML form, including node types, node parameters, connections and conditional expressions.
When the Agent queries order detail data (old search detail), the data is returned in JSON format.
To obtain a meaningful view of changes, it is usually necessary to first parse legal text into a structured format (such as XML or JSON) and then compare differences at the structural level, rather than directly running diff on plain text.
The tutorial uses a small model on the order of Qwen 3 6B as the baseline, outputting JSON for suspicious credit card transactions, with a schema containing three fields: priority (high/low), decision (fraud/fraud review/clear) and escalate (boolean).
Developers can send and receive JSON-format battle messages (such as `|request|` and `|turn|`) according to the Pokémon Showdown protocol, allowing Agents to compete against humans or other bots.
More efficient AI-native interaction should directly consume structured data (such as JSON, API responses, DOM trees), or directly schedule tasks through an intent protocol, thereby bypassing the intermediate layer of rendering and recognition.
Casey Muratori advocates never using JSON to store game data, citing JSON's slow parsing, lack of type validation, poor editing experience, and large amounts of runtime garbage, and recommends binary formats plus custom editors.
Having a model output JSON in free text is an extremely fragile approach; one extra word or one missing bracket in the output causes parsing to fail.
The core mechanism of URL-encoded documents typically relies on serialization formats (such as JSON) and compression algorithms: the document state is converted to a string and compressed with LZ-series algorithms (such as LZString) to reduce size, then embedded into the URL via Base64 or percent-encoding.
In the benchmark on the official SerpApi blog, the same seven search types (Search, Finance, News, Maps, Flights, Hotels, Shopping) were sent as JSON and as Markdown respectively; the JSON versions totaled 4668 tokens and the Markdown versions 3412 tokens, a 27% reduction.
40 more timeline events
All Facts (20)
The core mechanism of Function Calling is that developers describe external function signatures in JSON Schema format, and the model outputs structured JSON invocation instructions that conform to the schema specification.
80%VerifiedThe basic principle of tool calling is that the model identifies scenarios requiring external capabilities, outputs a structured function call request (usually in JSON format), and after the external system executes it, the result is returned to the model to continue reasoning.
75%VerifiedFunction Call does not actually execute code; the model only outputs a structured JSON string, and the actual execution is performed by the host program.
75%VerifiedProtocol Buffers reduces size by approximately 60–80% and parses 3–10 times faster compared to JSON.
75%VerifiedJSON organizes information into key-value pairs that flow through n8n workflow connections.
70%VerifiedSwitching to the Input tab in the n8n node editing panel allows you to preview in real time the complete JSON structure flowing into that node
65%VerifiedIn the tool-calling mechanism, developers attach a tool description list in JSON Schema format when calling the LLM API, and the model outputs call parameters in structured JSON format
65%VerifiedThe underlying implementation of tool calling typically relies on the Function Calling protocol, where developers describe a tool's name, parameter types, and semantics in JSON Schema format, and the model outputs structured invocation instructions conforming to the protocol format.
65%Verifiedn8n's underlying architecture is built on Node.js, workflow data is stored in JSON format, and the execution engine supports both serial and parallel task scheduling.
65%VerifiedThe JSON specification (RFC 8259) does not allow trailing commas, unlike JavaScript objects.
65%VerifiedEach Skill declares its own functional description, API endpoint, authentication method, and parameter specifications through a configuration file (typically in YAML or JSON format).
65%VerifiedWeChat Cloud Database stores data in JSON format, with no need to predefine table structures
65%UnverifiedCompared to JSON, protobuf has a smaller encoding size and faster serialization/deserialization speed.
70%UnverifiedJSON相比XML或Protocol Buffers，在LLM提示工程中的可读性和token效率都更优
70%UnverifiedIn the four-layer partition architecture, the second layer—structured profiles—uses structured JSON objects as the Single Source of Truth (SSOT) and employs a state-overwrite mechanism to handle data updates.
60%UnverifiedIn a workflow platform, the DSL typically defines the flow structure in JSON or YAML form, including node types, node parameters, connections and conditional expressions.
50%UnverifiedWhen the Agent queries order detail data (old search detail), the data is returned in JSON format.
50%UnverifiedTo obtain a meaningful view of changes, it is usually necessary to first parse legal text into a structured format (such as XML or JSON) and then compare differences at the structural level, rather than directly running diff on plain text.
50%UnverifiedThe tutorial uses a small model on the order of Qwen 3 6B as the baseline, outputting JSON for suspicious credit card transactions, with a schema containing three fields: priority (high/low), decision (fraud/fraud review/clear) and escalate (boolean).
50%UnverifiedDevelopers can send and receive JSON-format battle messages (such as `|request|` and `|turn|`) according to the Pokémon Showdown protocol, allowing Agents to compete against humans or other bots.
50%