Atomic Agent is a local-first AI agent that runs its control loop and state on a user's machine while using local or cloud models. It drives browsers, edits files, runs approved commands, remembers context, schedules follow-ups, and connects to external tools, with the catalogue providing skills for its use.
Borrowing it
Nothing to install: this file belongs to AtomicBot-ai/atomic-agent. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/AtomicBot-ai/atomic-agent/main/AGENTS.mdgit clone --depth 1 https://github.com/AtomicBot-ai/atomic-agentWrote 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.
[](https://agentmods.dev/instructions/atomicbot-ai/atomic-agent/agents-md)<a href="https://agentmods.dev/instructions/atomicbot-ai/atomic-agent/agents-md"><img src="https://agentmods.dev/badge/instructions/atomicbot-ai/atomic-agent/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.
<a href="https://agentmods.dev/instructions/atomicbot-ai/atomic-agent/agents-md"><img src="https://agentmods.dev/badge/instructions/atomicbot-ai/atomic-agent/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.01566 | $0.01566 |
| Opus 5.5 | $0.00626 | $0.00626 |
| Sonnet 5.5 | $0.00313 | $0.00313 |
| Haiku 4.5 | $0.00157 | $0.00157 |
Grade A, and why
atomic-agent 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 9d 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 — 80 lines — stays where its author put it; the contents beside it link to each section on GitHub.
atomic-agent: contributor instructions
This is the repository-wide contract. Detailed implementation documents are read only for the task at hand. User working agreements still apply.
Purpose and invariants
- A local operator runtime with CLI, TUI and NDJSON sidecar. Inference connects to external local/cloud providers; model/server/browser binaries are separate assets.
- Session state, durable memory, world snapshots and compressed observations live outside the model. The prompt is a bounded slice of that state.
- Keep stable-prefix bytes stable within the session/role scope. Put step-mutable data after conversation; per-request grammar does not belong in the cached prefix.
- One inference per step. Production grammar emits an array, even for one tool; runtime drives tool execution and subsequent steps.
- Grammar, prompt profile, stream parser and transport must agree. Keep reasoning bounds and terminal-step variants.
- A session is multi-turn chat: reply ends a turn, finish ends the session. Public frontends use runtime.runTurn; the sidecar's already-locked executeTurn seam must not enqueue recursively.
- Pass dependencies explicitly. getConfig is the established cache exception; preserve documented lifecycle exceptions rather than inventing new global state.
- Keep approval, read scope, input-file protection and cancellation enforced at dispatch. Never bypass them through prompt wording.
Working in an area
Before editing, read the local AGENTS.md for each touched area and for both sides of an interface change. Read its README, then only the linked mechanism documents relevant to the task. Do not assume every client loads nested instructions from the root automatically. Code and tests establish shipped behavior; archive/proposal documents do not override it.
agent/: agent instructions; module guide.runtime/: runtime instructions; module guide.config/: config instructions; module guide.prompt/: prompt instructions; module guide.session/: session instructions; module guide.llm/: llm instructions; module guide.local-llm/: local-llm instructions; module guide.memory/: memory instructions; module guide.tools/: tools instructions; module guide.tui/: tui instructions; module guide.channels/: channels instructions; module guide.mcp/: mcp instructions; module guide.skills/: skills instructions; module guide.tasks/: tasks instructions; module guide.http/: http instructions; module guide.tracing/: tracing instructions; module guide.approval/: approval instructions; module guide.scripts/: instructions; packaging/checks.
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.
- 9d ago First seen · 80 lines · 1,566 tokens per session scan A be7c164b4a2c
atomic-agent AGENTS.md is an instructions file published in the GitHub repository AtomicBot-ai/atomic-agent (3,347 stars, last pushed yesterday), licensed MIT. It adds 1,566 tokens to every session, about $0.0063 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 2026-10-07.
Other instructions, from other repositories
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
deepseek-harness AGENTS.md
AGENTS.md instructions for deepseek-ai/deepseek-harness, covering agents.md, pre-stable apis and released session data, repository layout, commands and host sandbox failures.
vscode copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).