domain

Repository guidance for using domain documentation in the K8E project. Domain documentation describes the project's decisions and terminology, while KIPs are its numbered architecture proposals.

In plain words
What is it for?
Use it when investigating a feature, checking whether an architecture proposal is implemented, or locating code for the sandbox command-line tools, clients, and gateway. It also points to the project's overview and glossary files.
Why use it?
It tells engineering agents which documents are authoritative and what to read before exploring or changing code. This helps prevent work based on outdated decisions or inconsistent terminology.

Agent

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 agents/xiaods/k8e/domain
Clone the repo
git clone --depth 1 https://github.com/xiaods/k8e
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 916 The whole file, excluding the scripts and references it only reads on demand.
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 $0.00000 $0.00916
Opus 5 $0.00000 $0.00458
Sonnet 5 $0.00000 $0.00183
Haiku 4.5 $0.00000 $0.00092

Measured yesterday against content hash d82c2919264b, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

domain 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 yesterday.

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.

docs/agents/domain.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.

Domain Docs

How the engineering skills should consume this repo's domain documentation.

Before exploring, read these

  • docs/README.md — KIP index and status. This is the source of truth for which proposals are Implemented, Partially implemented, or Outdated.
  • The KIP that touches the area you're working in (KIP-1 … KIP-25). Status on the KIP header must match the index; if they disagree, trust the index and fix the header.
  • CLAUDE.md / AGENTS.md at the repo root — project overview, architecture, build/test instructions.
  • CONTEXT.md — if it exists in the future, read it for domain language and glossary.

File structure

/
├── CLAUDE.md / AGENTS.md        # Project overview + architecture
├── docs/
│   ├── README.md                # KIP index + status (read this first)
│   ├── kip-1-…kip-25-*.md       # Architectural Decision Records
│   ├── agents/                  # Skill conventions (this file; not KIPs)
│   └── assets/
├── pkg/
│   ├── sandboxcli/              # CLI command handlers (k8e-sandbox-cli)
│   ├── sandbox/client/          # gRPC client + skill installation
│   ├── sandbox/e2b/             # E2B-compatible HTTP/Connect layer (KIP-18)
│   ├── sandboxmatrix/           # Core sandbox orchestration (warm pools, sessions)
│   ├── sandboxlayer/            # CAS snapshot layerstore (KIP-16 M2)
│   ├── server/                  # Control plane daemon
│   ├── agent/                   # Agent daemon (kubelet + containerd)
│   └── ...
├── pkg/sandboxcli/skills/k8e-sandbox/  # Embedded SKILL.md (installed via connect)
├── plugins/deepseek-harness/    # dsh-k8e-sandbox plugin family (KIP-20)
├── sandbox/                     # Sandbox container runtime shim
└── sandboxd/                    # Runtime daemon (Zig)

Do not treat KIP-4 / KIP-5 as current design: they are Outdated (MCP). There is no docs/sandbox-mcp-quickstart.md and no pkg/sandboxmcp/.

Key terms

When naming domain concepts (in issue titles, PRDs, refactors), use the terms as defined here:

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. yesterday First seen · 59 lines · 0 tokens per session scan A d82c2919264b

Subscribe to this mod's changes

domain is an agent published in the GitHub repository xiaods/k8e (480 stars, last pushed 7d ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 916 tokens. 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.