k8e: Instructions file for Codex

AGENTS.md

k8e AGENTS.md is an instructions file for Codex, OpenCode from xiaods/k8e. It costs 2,427 tokens per session, scanned A, original, Apache-2.0.

Repository guidance for Codex, an AI coding agent, working on K8E, a Kubernetes distribution for running AI agents in isolated environments. It explains the project structure, architecture, key workflows, and build process.

In plain words
What is it for?
Understanding K8E’s components, tracing how sandbox operations work, navigating the Go and Zig code, and building or modifying its server, agent, command-line tools, gateway, and software libraries.
Why use it?
It gives the coding agent the project context needed to make changes in the right places and follow the repository’s design and build conventions.

Instructions file for CodexOpenCode

Written for Codex and OpenCode: the file is AGENTS.md. Also seen: reads .claude/ paths; mentions CLAUDE.md; mentions Claude Code.

This is xiaods/k8e's own configuration. It tells Codex and OpenCode how to work on k8e itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything k8e configures →

Reuse

Borrowing it

Nothing to install: this file belongs to xiaods/k8e. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/xiaods/k8e/main/AGENTS.md
Clone the repo
git clone --depth 1 https://github.com/xiaods/k8e

Made for: Codex, OpenCode.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for k8e AGENTS.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/xiaods/k8e/agents-md/github.svg)](https://agentmods.dev/instructions/xiaods/k8e/agents-md)
Your own site
<a href="https://agentmods.dev/instructions/xiaods/k8e/agents-md"><img src="https://agentmods.dev/badge/instructions/xiaods/k8e/agents-md/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for k8e AGENTS.md

Your own site · 80×15
<a href="https://agentmods.dev/instructions/xiaods/k8e/agents-md"><img src="https://agentmods.dev/badge/instructions/xiaods/k8e/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 2,427 This file is loaded in full into every session.
When invoked 2,427 The same file — it is already loaded in full.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.02427 $0.02427
Opus 5.5 $0.00971 $0.00971
Sonnet 5.5 $0.00485 $0.00485
Haiku 4.5 $0.00243 $0.00243

Measured 20d ago against content hash 1fed32d491b0, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

k8e AGENTS.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 20d 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.

AGENTS.md · 192 lines

How it starts

The opening of the file, as written. The whole thing — 192 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AGENTS.md

Guidance for AI coding agents (Claude Code, Codex, and others) working in this repository.

Claude Code reads this file directly — there is no separate CLAUDE.md. Keep AGENTS.md as the single source of agent guidance.

What this is

K8E is a CNCF-conformant Kubernetes distribution packaged as a single k8e binary under 100MB. Kubernetes is the substrate; the product on top of it is a sandbox matrix for isolated AI-agent execution:

  • warm-pooled sandbox Pods — RuntimeClass gvisor (default), kata, firecracker
  • a gRPC SandboxService (34 RPCs) behind sandbox-grpc-gateway:50051 with mTLS
  • k8e-sandbox-cli — the CLI an agent drives from a shell (30 top-level commands)
  • an E2B-compatible HTTP surface, an MCP adapter, and skills/plugins for Claude Code, Codex, Pi and DeepSeek Harness (dsh)

There is no k8e sandbox command group. The k8e binary carries only k8e sandbox-apikey and k8e e2b-server; every session/exec/file command lives in k8e-sandbox-cli.

Where things live

The build is a Zig build graph (build.zig) driving Go compilation; Makefile is a thin wrapper over the common zig build steps.

.
├── build.zig / Makefile     # build graph / `make` wrappers
├── main.go                  # CLI entry; holds the //go:generate directives
├── cmd/                     # one package per binary entry point
│   ├── server/              # → bin/k8e (control plane + agent, no bundled runtimes)
│   ├── k8e/                 # release multicall binary (hack/package-cli)
│   ├── sandboxcli/          # → k8e-sandbox-cli, the agent-facing CLI
│   ├── agent/ kubectl/ cert/ ctr/ encrypt/ etcdsnapshot/ token/ …
├── pkg/
│   ├── cli/cmds/            # flag + command wiring (server.go, agent.go, e2b_server.go)
│   ├── server/ agent/ daemons/
│   ├── sandboxmatrix/       # the core: warm-pool + session reconciliation,
│   │                        #   leader.go (Lease), ratelimit/, api/v1alpha1 CRDs,
│   │                        #   grpc/ (SandboxService impl + orchestrator + pb/)
│   ├── sandboxcli/          # the 30 top-level commands + skills/k8e-sandbox/SKILL.md
│   ├── sandbox/             # client/ (gRPC + TLS discovery), apikey/, e2b/
│   ├── sandboxmcp/          # MCP HTTP adapter (`mcp-serve`)
│   ├── sandboxlayer/        # CAS snapshot layerstore
│   ├── configfilearg/       # maps config-file keys onto flags
│   ├── deploy/ embedw/
│   └── …                    # apis, bootstrap, token, certmonitor, cgroups
├── proto/sandbox/v1/sandbox.proto   # SandboxService: 34 RPCs
├── sandboxd/                # Zig daemon running inside the sandbox Pod (:2024)
├── sandbox/ manifests/ plugins/deepseek-harness/
├── docs/                    # KIPs (docs/README.md = index) + docs/agents/
├── hack/                    # download / generate / package / e2e scripts
└── tests/                   # unit.go helpers + mcp/*.py smoke tests

Read the full file on GitHub · 192 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. 20d ago Changed · +2 tokens per session 1fed32d491b0
  2. 1mo ago Changed · -54 lines · -1,056 tokens per session 59eef4870a69
  3. 1mo ago Changed · +104 lines · +1,790 tokens per session 4b394e3b6c01
  4. 2mo ago First seen · 142 lines · 1,691 tokens per session scan A 45cb3ad812af

Subscribe to this mod's changes

k8e AGENTS.md is an instructions file published in the GitHub repository xiaods/k8e (514 stars, last pushed 10d ago), licensed Apache-2.0. It adds 2,427 tokens to every session, about $0.0097 per session on Opus 5.5. 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

k8e AGENTS.md compared with instructions from other repositories

k8e AGENTS.md next to 6 instructions from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
k8e AGENTS.mdxiaods/k8eInstructions file51410d agoApache-2.0A2,427Codex, OpenCode
vscode buildNext.instructions.mdmicrosoft/vscodeInstructions file195kyesterdayMITA9,719GitHub Copilot
next.js AGENTS.mdvercel/next.jsInstructions file143ktodayMITA7,461Codex, OpenCode
langchain AGENTS.mdlangchain-ai/langchainInstructions file149kyesterdayMITA4,680Codex, OpenCode
vscode oss-third-party-notices.instructions.mdmicrosoft/vscodeInstructions file195kyesterdayMITA5,516GitHub Copilot
vscode copilot-instructions.mdmicrosoft/vscodeInstructions file195kyesterdayMITA3,093GitHub Copilot
deepseek-harness AGENTS.mddeepseek-ai/deepseek-harnessInstructions file274k2d agoMITA4,313Codex, OpenCode