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.
npx agentmods add instructions/noontide-co/mainbranch/agents-mdgit clone --depth 1 https://github.com/noontide-co/mainbranchWrote 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/noontide-co/mainbranch/agents-md)<a href="https://agentmods.dev/instructions/noontide-co/mainbranch/agents-md"><img src="https://agentmods.dev/badge/instructions/noontide-co/mainbranch/agents-md.svg" alt="Measured on agentmods" 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 | $0.04464 | $0.04464 |
| Opus 5 | $0.02232 | $0.02232 |
| Sonnet 5 | $0.00893 | $0.00893 |
| Haiku 4.5 | $0.00446 | $0.00446 |
Grade A, and why
mainbranch 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 5d 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 — 530 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Instructions
Main Branch is public open-source infrastructure. Treat this file as the repo-level operating contract for Codex, Claude Code, and any other agent or human contributor working in this repository.
Product Shape
Main Branch is a CLI plus agent-runtime skill system for running a business from markdown files in git.
mbis the deterministic, inspectable, scriptable control plane.- Agent-runtime skills are the judgment-heavy execution layer.
- Shared workflow sources are the runtime-agnostic workflow contract; runtime shells and generated guidance are adapters over those sources.
- GitHub issues are tasks.
- Pull requests are proposals and review conversations.
- Git history is the evolution story.
- A user's business repo is the durable business brain.
- A user is not met with git speak but instead business speak.
Claude Code and Codex CLI are the supported agent runtimes today. Claude Code
uses project-local slash skills. Codex uses generated AGENTS.md, global Main
Branch skills, deterministic mb facts, and workflow inventory. That is
daily-loop Codex support, not full production workflow parity. Business repos
should not carry repo-local Codex plugin or skill copies unless a future issue
proves that fallback is required.
Cursor, OpenClaw, Hermes, Paperclip-adjacent orchestration, and local runtimes
remain compatibility targets until tested. Do not claim support before there is
an adapter and smoke evidence for the exact surface.
The public product frame lives in docs/ethos.md, the operator loop taxonomy
lives in docs/operator-loops.md, and release direction lives in
docs/roadmap.md.
Daily Operating Loop
The current product center is the daily owner loop:
- The operator opens the business repo and starts Claude Code or Codex CLI.
/mb-start, the generated repo instructions, or global Main Branch Codex skills ground the agent in deterministicmbfacts before advice.- The agent routes thought dumps and requests into business primitives: bets, goals, offers, research, decisions, pushes, playbooks, outcomes, and checkpoints.
mbkeeps the repo healthy: repo shape, status health, graph links, provider readiness, runtime wiring, update/repair paths, validation, and guarded commits./mb-endandmb checkpointreconcile what changed so durable business memory stays in git.
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.
- 5d ago First seen · 530 lines · 4,464 tokens per session scan A 2bbfc46b50b7
mainbranch AGENTS.md is an instructions file published in the GitHub repository noontide-co/mainbranch (38 stars, last pushed 21d ago), licensed MIT. It adds 4,464 tokens to every session, about $0.0223 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.
Other instructions, from other repositories
shieldcn AGENTS.md
AGENTS.md instructions for jal-co/shieldcn, covering agents.md — shieldcn, what this is, monorepo structure, packages/core (@shieldcn/core) and packages/web (@shieldcn/web).
markdown-site CLAUDE.md
Instructions for waynesutton/markdown-site, covering claude.md, project context, quick start, commands and default and legacy modes.
markdown-site AGENTS.md
Instructions for waynesutton/markdown-site, covering agents.md, project overview, default and legacy modes, current status and tech stack.
Claw3D AGENTS.md
Instructions for iamlukethedev/Claw3D, covering agent instructions, cursor cloud specific instructions, service overview, running the app and lint, typecheck, and tests.
artifacts CLAUDE.md
Claude Code instructions for dbtlr/artifacts, a project described as: Persistent links for agent-generated Markdown, HTML, and text via MCP.
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).