# JSON-RPC Projects at AI Tinkerers

> Canonical HTML: https://aitinkerers.org/technologies/json-rpc
> Markdown URL: https://aitinkerers.org/technologies/json-rpc.md
> Technology record last updated: 2026-04-13T14:18:05Z
> Generated: 2026-10-09T04:18:25Z

JSON-RPC is a stateless, lightweight Remote Procedure Call (RPC) protocol, using JSON exclusively to encode method invocation requests and responses.

JSON-RPC (version 2.0 is standard) defines a simple, transport-agnostic protocol for remote service interaction: it is not bound to HTTP, operating over sockets, WebSockets, or other streams. A client sends a Request object—containing the protocol version (`"jsonrpc": "2.0"`), the remote function name (`"method": "getBalance"`), parameters (`"params": ["0x123..."],` an array or object), and a unique `id`—to the server. The server executes the method and returns a Response object with the matching `id` and either a `result` or an `error` object. This structure facilitates efficient batch requests and one-way notifications (requests without an `id`), making it a preferred choice for high-throughput systems, notably in the blockchain ecosystem (e.g., Ethereum's API).

- Official technology site: https://www.jsonrpc.org/specification
- Public AI Tinkerers demos and talks: 3
- Result page: 1 of 1

## Recent Public Talks and Demos

### [Twine: Make Coding Agents Challenge Each Other](https://seattle.aitinkerers.org/talks/rsvp_AHtybiIbQsM)

I built Twine, a local desktop application that coordinates Codex CLI, Claude Code and pi as independent coding sessions in one workspace. Twine passes tasks and findings between sessions automatically, so the user does not have to copy context between open terminal windows. Each agent keeps its native terminal and conversation. Twine gives one participant write control at a time and records assignments, handoffs, reports, traces and changed artifacts without scraping terminal text to guess when a turn ended. Twine can put a higher-end model such as Fable or Astra in Main's outer loop to delegate, check progress on a cadence and intervene when needed. Self-hosted models can handle the repeated build and review turns. That can reduce paid-model token use and avoids provider usage quotas for those local-model turns, while still using local compute. The user can inspect and steer every session. In the five-minute demo, Codex CLI with gpt-5.6-sol runs Main, Qwen3.8 Flash Next on an NVIDIA DGX Spark builds a small HTML page and Claude Opus 5 reviews it. I will spend 40 seconds on setup and 45 seconds on the live Main and Qwen terminals. The next two minutes and 20 seconds cover the retained run: Claude's browser-based review, which found two accessibility bugs, the graph handoff and Qwen's reciprocal audit. I will use the final 75 seconds for Main's verification, the corrected page and the reusable pattern: hand off on native session-completion events while allowing only one writable turn.

- Event context: AI Dev Tools Track - Seattle - October 1 — 2026-10-02 — Seattle
- Public talk page: https://seattle.aitinkerers.org/talks/rsvp_AHtybiIbQsM

### [CodeDB: Building a Code Intelligence Server That Cuts Agent Token Usage](https://singapore.aitinkerers.org/talks/rsvp_YyFDIYV6hPs)

CodeDB is a code intelligence server for AI agents that indexes a codebase once and serves structured answers through MCP, HTTP, and CLI instead of forcing models to repeatedly scan raw files. In this demo, I show how CodeDB uses structural indexing, trigram search, inverted word lookups, dependency graphs, and snapshots to help agents find only the most relevant code context with much lower latency and dramatically fewer tokens.

- Event context: AI Tinkerers Singapore: The Agentic Future & Dev/Eng Workflows — 2026-04-21 — Singapore
- Public talk page: https://singapore.aitinkerers.org/talks/rsvp_YyFDIYV6hPs

### [A2A Deep dive](https://dublin.aitinkerers.org/talks/rsvp_PUfCKEuGrfQ)

Agent To Agent (A2A) is a newly proposed protocol by google that aims at standardizing how Agents communicate with other vendors, this is particularly useful in the case of 3rd party AI Agents. In this demo, I'll walk through some of the key design principles of A2A, by walking through code and highlighting how the principles are/can be implemented. We'll step through what a client and server for A2A should look like and how communication between the two is facilitated by a task manager.

- Event context: AI Tinkerers - Dublin Event (April) — 2025-04-24 — Dublin
- Public talk page: https://dublin.aitinkerers.org/talks/rsvp_PUfCKEuGrfQ

## Related Technologies

- [MCP](https://aitinkerers.org/technologies/mcp) ([Markdown](https://aitinkerers.org/technologies/mcp.md)) — 133 public demos
- [Codex CLI](https://aitinkerers.org/technologies/codex-cli) ([Markdown](https://aitinkerers.org/technologies/codex-cli.md)) — 6 public demos
- [crew](https://aitinkerers.org/technologies/crew) ([Markdown](https://aitinkerers.org/technologies/crew.md)) — 1 public demo
- [dependency graphs](https://aitinkerers.org/technologies/dependency-graphs) ([Markdown](https://aitinkerers.org/technologies/dependency-graphs.md)) — 1 public demo
- [GitHub](https://aitinkerers.org/technologies/github) ([Markdown](https://aitinkerers.org/technologies/github.md)) — 76 public demos
- [Gradio](https://aitinkerers.org/technologies/gradio) ([Markdown](https://aitinkerers.org/technologies/gradio.md)) — 9 public demos
- [headless Chrome](https://aitinkerers.org/technologies/headless-chrome) ([Markdown](https://aitinkerers.org/technologies/headless-chrome.md)) — 1 public demo
- [HTTP](https://aitinkerers.org/technologies/http) ([Markdown](https://aitinkerers.org/technologies/http.md)) — 14 public demos
- [HTTP/1](https://aitinkerers.org/technologies/http-1) ([Markdown](https://aitinkerers.org/technologies/http-1.md)) — 1 public demo
- [inverted word indexing](https://aitinkerers.org/technologies/inverted-word-indexing) ([Markdown](https://aitinkerers.org/technologies/inverted-word-indexing.md)) — 1 public demo
- [OpenAI compatible model](https://aitinkerers.org/technologies/openai-compatible-model) ([Markdown](https://aitinkerers.org/technologies/openai-compatible-model.md)) — 1 public demo
- [Python](https://aitinkerers.org/technologies/python) ([Markdown](https://aitinkerers.org/technologies/python.md)) — 673 public demos
- [structural parsing](https://aitinkerers.org/technologies/structural-parsing) ([Markdown](https://aitinkerers.org/technologies/structural-parsing.md)) — 1 public demo
- [trigram indexing](https://aitinkerers.org/technologies/trigram-indexing) ([Markdown](https://aitinkerers.org/technologies/trigram-indexing.md)) — 1 public demo
- [Zig](https://aitinkerers.org/technologies/zig) ([Markdown](https://aitinkerers.org/technologies/zig.md)) — 4 public demos