xbcode CLAUDE.md

Repository instructions for xbcode, a TypeScript command-line program with a terminal interface and an AI agent loop. They describe its commands, file structure, API modes, tools, and environment settings.

In plain words
What is it for?
Use them when developing, building, starting, or debugging xbcode and its file and task tools.
Why use it?
They give a coding agent the project context needed to make changes that fit the existing design and run the right commands.

Instructions file

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add instructions/lwmxiaobei/xbcode/claude-md
Clone the repo
git clone --depth 1 https://github.com/lwmxiaobei/xbcode
Per session 1,116 This file is loaded in full into every session.
When invoked 1,116 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.01116 $0.01116
Opus 5 $0.00558 $0.00558
Sonnet 5 $0.00223 $0.00223
Haiku 4.5 $0.00112 $0.00112

Measured 2d ago against content hash ce8cc50e1f57, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

xbcode CLAUDE.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

CLAUDE.md · 59 lines

How it starts

The opening of the file, as written. The whole thing — 59 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md

This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.

Project Overview

A TypeScript CLI agent (similar to Claude Code) built with OpenAI SDK + Ink (React for CLI). It implements an agent loop with tool use, streaming output, and a terminal UI.

Commands

npm run dev      # Run in development (tsx, no compile)
npm run build    # Compile TypeScript to dist/
npm start        # Run compiled output

Architecture

Multi-file architecture: src/index.tsx (UI), src/agent.ts (agent loop), src/agent/ (loop internals: streaming, tool calls, usage, messages), src/tools.ts (tool definitions & handler routing), src/tools/ (file tool implementations), src/types.ts.

Dual API mode: Supports both OpenAI Responses API (default) and Chat Completions API (for DeepSeek-compatible endpoints). Auto-detected via OPENAI_BASE_URL or set explicitly with OPENAI_API_MODE.

Agent loop: agentLoop() (Responses API) and agentLoopWithChatCompletions() (Chat Completions) implement the core think-act cycle. Both loop until LLM returns no tool calls.

Tools: Defined in src/tools.ts (schemas + handler routing). Beyond the file and shell tools there are glob/grep, MCP access, skills, web fetch/search, the task board, goals, teammates, and subagent (dispatches work to an independent sub-agent with a clean context; sub-agents get base tools only, no recursive dispatch). Tool schemas use JSON Schema with additionalProperties: false.

A handler returns either a plain string or { output, details }output is what the model sees, details carries structured data for the UI only (currently edit_file's diff and unified patch). executeToolCall normalizes both shapes.

File tools: implemented under src/tools/, not in tools.ts:

  • file-tools.tsread_file / write_file / edit_file, plus injectable operations so the tools can be pointed at a remote filesystem.
  • edit-diff.ts — matching and diff generation. edit_file takes an edits[] array applied atomically; every old_text is matched against the original file and must hit exactly one non-overlapping place, otherwise the call fails and nothing is written. Exact match is tried first, then a fuzzy pass (NFKC, per-line trailing whitespace, curly quotes, Unicode dashes/spaces); a fuzzy hit only rewrites the lines it touches so untouched bytes survive. BOM and CRLF are preserved.
  • file-truncate.tsread_file caps output at 2000 lines or 50KB, whichever hits first, and tells the model the exact offset to pass to continue.
  • file-mutation-queue.ts — serializes writes to the same file (keyed by realpath); different files still run in parallel.
  • image.ts — header-sniffed image support for read_file. Since neither API accepts images in a tool result, read_file queues the image on RunControl.pendingImages and the agent loop sends it as a follow-up user message.

Read the full file on GitHub · 59 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 59 lines · 1,116 tokens per session scan A ce8cc50e1f57

Subscribe to this mod's changes

xbcode CLAUDE.md is an instructions file published in the GitHub repository lwmxiaobei/xbcode (48 stars, last pushed 20d ago), licensed MIT. It adds 1,116 tokens to every session, about $0.0056 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other instructions, from other repositories

spec-kit AGENTS.md

AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.

github/spec-kit · 7,104 tokens

vscode buildNext.instructions.md

Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).

microsoft/vscode · 6,785 tokens

codex AGENTS.md

AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.

openai/codex · 5,182 tokens

langchain AGENTS.md

AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.

langchain-ai/langchain · 4,345 tokens

vscode oss-third-party-notices.instructions.md

Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).

microsoft/vscode · 5,001 tokens

next.js AGENTS.md

Instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.

vercel/next.js · 7,296 tokens