OpenCode TUI plugin that displays real-time LLM output token speed (tok/s)

Best-effort parser for JSON that hasn't finished arriving. Render streaming LLM output before the closing brace shows up. Zero dependencies, never throws.

Extract & validate JSON from messy LLM output. Designed for Claude / Codex CLI workflows where tool-use is unavailable. XML-tag aware, with multi-stage fallbacks and built-in repair.

Strip markdown fences and extract clean data from raw LLM output

Yomi (読み) - Flexible JSON parser that interprets LLM output to match Zod schemas

Lightweight prompt templating with {{var}} interpolation and version tracking, so you know which prompt version produced which LLM output.

MCP server that generates professional corporate PDF reports from structured content or raw LLM output

TypeScript SDK for the Nexara speech-to-text API: transcription, diarization, speaker roles, emotion recognition, structured LLM output, billing

Cloudflare Workers MCP server wrapper: llm-output-validator

one-line LLM output parsers for JS/TS

Lenient markdown → Telegram Bot API { text, entities } renderer built for LLM output: no parse_mode, no escaping, no parse errors; entity-aware message splitting, streaming-friendly partial rendering, offline preview.

Make LLM output 'look nicer'

Cloudflare Workers MCP server wrapper: llm-output-quality-monitor

Headless React hooks and primitives for displaying streamed LLM output well.

MCP server exposing llm-output-sanitizer: strip dangerous HTML, SQL, and shell snippets from LLM output before rendering, executing, or persisting. Built for Claude Desktop, Cursor, Cline, Windsurf, and Zed.

Semi-incremental markdown parser and renderer, optimized for LLM output.

Cloudflare Workers MCP server wrapper: llm-output-evaluator

Heuristic LLM output quality scoring without calling another LLM

Pre-commit for agents — validate LLM output before applying

Schema-aware JSON repair engine for LLM output recovery and user input validation