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/thrashr888/agentkernel/claude-mdgit clone --depth 1 https://github.com/thrashr888/agentkernelWhat 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 | $0.01590 | $0.01590 |
| Opus 5 | $0.00795 | $0.00795 |
| Sonnet 5 | $0.00318 | $0.00318 |
| Haiku 4.5 | $0.00159 | $0.00159 |
Grade A, and why
agentkernel 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 3d 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 — 183 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
agentkernel is a Firecracker-based microVM runtime for running AI coding agents (Claude Code, Gemini CLI, Codex, OpenCode) in true hardware-isolated sandboxes. It provides per-project isolation with sub-125ms boot times, ~25MB images, and real kernel-level isolation via KVM.
Key difference from containers: Each sandbox runs in its own virtual machine with a dedicated Linux kernel, not a shared kernel like Docker. This provides stronger isolation guarantees.
Build Commands
# Build
cargo build
# Build release
cargo build --release
# Run tests
cargo test
# Run stress test (100 VMs, requires Firecracker implementation)
cargo test --test stress_test -- --nocapture --ignored
# Quality gates (run before commits)
cargo fmt -- --check && cargo clippy -- -D warnings && cargo test
# Run the CLI
cargo run -- <command>
Building VM Images
# Build kernel (via Docker, works on macOS)
cd images/build
docker build -t agentkernel-kernel-builder -f Dockerfile.kernel-builder .
docker run --rm -v "$(pwd)/../kernel:/kernel" agentkernel-kernel-builder 6.18.45
# Output: images/kernel/vmlinux-6.18.45-agentkernel (~4MB)
Architecture
src/
├── main.rs # CLI entry point (clap-based), command dispatch
├── config.rs # Config parsing for agentkernel.toml
├── permissions.rs # Security profiles and permission management
├── agents.rs # Multi-agent support (Claude, Gemini, Codex, OpenCode)
├── ssh.rs # SSH cert auth: CA/client keygen, sshd config, cert signing
├── http_api.rs # HTTP REST API server (includes IP in responses)
├── mcp.rs # MCP server for Claude Code integration
├── docker_backend.rs # Docker/Podman container backend
├── vmm.rs # Virtual machine manager (abstracts backends)
├── languages.rs # Language/runtime detection
└── setup.rs # Setup and installation management
images/
├── kernel/
│ ├── microvm.config # Minimal kernel config for Firecracker
│ └── vmlinux-*-agentkernel # Built kernel (after build)
├── rootfs/ # Rootfs images (TODO)
└── build/
├── build-kernel.sh # Kernel build script
└── Dockerfile.kernel-builder # Docker build environment
tests/
└── stress_test.rs # 100 VM parallel stress test
plan/
└── firecracker-pivot.md # Full architecture plan
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.
- 3d ago First seen · 183 lines · 1,590 tokens per session scan A 69ddef3a7862
agentkernel CLAUDE.md is an instructions file published in the GitHub repository thrashr888/agentkernel (58 stars, last pushed 4d ago), licensed MIT. It adds 1,590 tokens to every session, about $0.0080 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.
Other instructions, from other repositories
ii-agent AGENTS.md
Instructions for Intelligent-Internet/ii-agent, covering ii-agent contributor guide, quick start, repository map, mandatory rules and architecture.
intelligent-terminal rust.instructions.md
Concise Rust coding conventions for this repository.
InvestSkill GEMINI.md
Instructions for yennanliu/InvestSkill, covering investskill — gemini cli setup & usage guide, installation & setup, quick start, navigate to the investskill directory and start gemini cli (loads gemini.md automatically).
sortie copilot-instructions.md
Instructions for sortie-ai/sortie, covering sortie coding & review standards, 1. layered imports (downward only; violation is critical), 2. concurrency safety, 3. workspace path safety (critical: security boundary) and 4. persistence (sqlite).
crust AGENTS.md
Instructions for chenxin-yan/crust, covering tests, documentation, changesets, agent skills and issue tracker.
she-love-me CLAUDE.md
Instructions for 863401402/she-love-me, covering claude.md and 唯一工作流.