oneagent AGENTS.md

Instructions for oneagent, a configurable command-line tool and Go library that gives several AI coding agents one common interface.

In plain words
What is it for?
Use it when developing, configuring, embedding, or documenting oneagent and its support for Claude, Codex, OpenCode, and Pi.
Why use it?
It explains the project’s commands, configuration, output handling, and reusable conversation storage in one place.

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

Made for: Codex, OpenCode.

Per session 540 This file is loaded in full into every session.
When invoked 540 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 $0.00540 $0.00540
Opus 5 $0.00270 $0.00270
Sonnet 5 $0.00108 $0.00108
Haiku 4.5 $0.00054 $0.00054

Measured 2d ago against content hash e98f52b73621, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

oneagent 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 · 42 lines

How it starts

The opening of the file, as written. The whole thing — 42 lines — stays where its author put it; the contents beside it link to each section on GitHub.

oneagent

Config-driven multi-agent CLI. One interface for Claude, Codex, OpenCode, Pi, and any future agent.

What this is

A Go library and CLI that wrap AI agent CLIs behind one normalized interface. Built-in backend defaults ship for Claude, Codex, OpenCode, and Pi, with optional overrides in ~/.config/oneagent/backends.json. New backends are added by config, not by code.

Repo: https://github.com/1broseidon/oneagent

Docs

  • README.md — landing page and CLI quick start
  • docs/library.md — detailed Go embedding guide
  • docs/config.md — backend config schema and examples
  • docs/examples/consumer.md — concrete consumer example

Architecture

  • oneagent.go — core runtime. Defines Client, builds backend commands, executes them, normalizes JSON and JSONL output, and emits streaming events.
  • config.go — backend config compiler. Turns the compact JSON config schema into canonical Backend structs.
  • thread.go — portable thread storage, native-session reuse, cross-backend replay, thread compaction, and the pluggable thread-store interface/default filesystem store.
  • cmd/oa/main.go — thin CLI wrapper. Defaults to plain text output, with --json for final machine output and --stream --json or --jsonl for normalized JSONL events.
  • Template variables: {prompt}, {model}, {cwd}, {session} — substituted into command arrays. Empty variables drop themselves and their preceding flag.
  • Two output formats: json (single blob) and jsonl (line-delimited events with configurable activity/delta/result/session/error selectors).
  • Match conditions support dot-paths and & for AND: type=message_update&assistantMessageEvent.type=text_delta
  • The normalized stream is intentionally small: session, optional activity, optional delta, then done or error.
  • Activity templates can interpolate JSON fields with {...} placeholders, including array indexes like {message.content.0.name}.
  • Package-level helpers remain backward-compatible wrappers over a default Client and filesystem-backed thread store.

Read the full file on GitHub · 42 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 · 42 lines · 540 tokens per session scan A e98f52b73621

Subscribe to this mod's changes

oneagent AGENTS.md is an instructions file published in the GitHub repository 1broseidon/oneagent (5 stars, last pushed 5mo ago), licensed MIT. It adds 540 tokens to every session, about $0.0027 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.