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.
npx agentmods add instructions/sudoitir/claudecode4j/claude-mdgit clone --depth 1 https://github.com/sudoitir/claudecode4jWhat 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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.01651 | $0.01651 |
| Opus 5 | $0.00826 | $0.00826 |
| Sonnet 5 | $0.00330 | $0.00330 |
| Haiku 4.5 | $0.00165 | $0.00165 |
Grade A, and why
claudecode4j 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.
How it starts
The opening of the file, as written. The whole thing — 182 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
ClaudeCode4J is a Java library for integrating with Claude Code CLI. It wraps the CLI tool and provides a clean Java API for executing prompts, streaming responses, and managing sessions.
Build Commands
# Build entire project
mvn clean install
# Build without tests
mvn clean install -DskipTests
# Run unit tests only (excludes integration and E2E tests)
mvn test
# Run a single test class
mvn test -Dtest=ClaudeRestIntegrationTest
# Run a single test method
mvn test -Dtest=ClaudeRestIntegrationTest#shouldExecutePromptAndReturnResponse
# Run tests for a specific module
mvn test -pl claudecode4j-core
# Run integration tests (requires Docker for Testcontainers)
mvn verify -Pintegration
# Run all tests including E2E (requires Claude CLI installed and authenticated)
mvn verify -Pall
# Format code (Spotless with Palantir Java Format)
mvn spotless:apply
# Check formatting
mvn spotless:check
Architecture
Module Structure
- claudecode4j-bom: Bill of Materials for version management
- claudecode4j-api: Interfaces, DTOs, and SPIs (no implementation)
- claudecode4j-core: Pure Java implementation (no Spring dependency)
- claudecode4j-spring-boot-starter: Spring Boot 4 auto-configuration
- claudecode4j-rest-adapter: REST/SSE endpoints with OpenAI and Anthropic API compatibility
- claudecode4j-kafka-adapter: Kafka request-reply messaging
- claudecode4j-websocket-adapter: WebSocket terminal for interactive sessions
- claudecode4j-mcp-server: MCP Server support for exposing Java methods as Claude tools
- claudecode4j-context: Token-aware context optimization using JTokkit
Key Design Patterns
Sealed Types: ClaudeResponse and ClaudeException use sealed interfaces/classes with pattern matching:
switch (response) {
case TextResponse text -> ...
case StreamResponse stream -> ...
case ErrorResponse error -> ...
}
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.
- 2d ago First seen · 182 lines · 1,651 tokens per session scan A 74ecdc1a5fcb
claudecode4j CLAUDE.md is an instructions file published in the GitHub repository sudoitir/claudecode4j (10 stars, last pushed 9d ago), licensed MIT. It adds 1,651 tokens to every session, about $0.0083 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 2026-08-31.
Other instructions, from other repositories
jikkou AGENTS.md
Instructions for streamthoughts/jikkou, covering agents.md - development guidelines for ai coding agents, project overview, project structure, build, test, and lint commands and build commands.
klag CLAUDE.md
Instructions for themoah/klag, covering claude.md, project overview, build commands, graalvm native image (startup/memory optimized) and architecture.
agent-skills CLAUDE.md
Instructions for confluentinc/agent-skills, covering claude.md, repository purpose, skill anatomy, evals are the contract and reviewing skill behavior.
kafka-backup CLAUDE.md
Instructions for osodevops/kafka-backup, covering project overview, key problems solved, repository structure, build & test commands and build the project.
rsyslog copilot-instructions.md
Copilot instructions for rsyslog/rsyslog, a project described as: High-performance log ingestion and ETL engine.
kafka-schema-reg-mcp AGENTS.md
Instructions for aywengo/kafka-schema-reg-mcp, covering development guidelines, prerequisites, quick setup, install dependencies and install dev dependencies (optional but recommended).