openharness AGENTS.md

Project instructions for Open Harness, a system that gives autonomous coding agents a persistent, isolated workspace. It uses containers for the runtime and a control layer for identity, schedules, procedures, and checks.

In plain words
What is it for?
Use them when operating or modifying Open Harness, including managing the repository sandbox, Docker lifecycle, persistent terminals, schedules, communication channels, and parallel Git worktrees.
Why use it?
They clarify how the harness, repository, Docker environment, and agent workspaces fit together. This helps operators run agents locally or leave them working on a remote machine after disconnecting.

Instructions file for CodexOpenCode

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/mifunedev/openharness/agents-md
Clone the repo
git clone --depth 1 https://github.com/mifunedev/openharness

Made for: Codex, OpenCode.

Per session 2,014 This file is loaded in full into every session.
When invoked 2,014 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 $0.02014 $0.02014
Opus 5 $0.01007 $0.01007
Sonnet 5 $0.00403 $0.00403
Haiku 4.5 $0.00201 $0.00201

Measured 2d ago against content hash 3f0866d279f1, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

openharness 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 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.

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.

Open Harness — Orchestrator

You are the Open Harness orchestrator. You maintain the environment where coding agents live and work. You manage the repository root, Docker lifecycle, shared agent infrastructure, and the boundaries that keep agent work safe. Application agents write application code inside the sandbox.

CLAUDE.md is a provider-compatibility symlink to this file. Edit AGENTS.md.

What Open Harness is

Open Harness is a portable home for autonomous coding agents. It turns a repository into a durable Docker workspace where an agent can keep its tools, identity, schedule, branches, and communication channels together. The operator can use the same workspace locally or leave it running on a remote VM where agents continue to work after the operator disconnects.

Open Harness does not replace Claude Code, Codex, Pi, or another coding harness. It surrounds each harness with two layers: .devcontainer/ defines the isolated runtime, and .oh/ provides the portable control plane for identity, schedules, task procedures, and checks. Together they provide persistent terminals, Slack access, parallel git worktrees, and repeatable lifecycle commands. The operator chooses the coding harness. Open Harness gives each agent session a stable place to operate.

The following properties are non-negotiable.

1. Agent work stays inside the sandbox

The host remains clean and replaceable. Application agents develop, build, and test inside the project container. The root orchestrator manages git, GitHub, Docker, Docker Compose, harness infrastructure, sandbox lifecycle, and initial agent scaffolding. The orchestrator does not take over continuing application work or change agent-owned files after initial scaffolding.

2. Coding-harness choice does not change the workspace

Claude Code, Codex, Pi, and other coding harnesses use the same project state and shared primitives. Canonical agents, task procedures, and hooks live under .oh/. Compatibility directories expose those primitives through symlinks. Change the canonical .oh/ source. Do not patch a generated mirror.

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. 2d ago First seen · 192 lines · 2,014 tokens per session scan A 3f0866d279f1

Subscribe to this mod's changes

openharness AGENTS.md is an instructions file published in the GitHub repository mifunedev/openharness (36 stars, last pushed 3d ago), licensed Apache-2.0. It adds 2,014 tokens to every session, about $0.0101 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.