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/basefoundry/base/agents-mdgit clone --depth 1 https://github.com/basefoundry/baseWhat 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 | $0.01086 | $0.01086 |
| Opus 5 | $0.00543 | $0.00543 |
| Sonnet 5 | $0.00217 | $0.00217 |
| Haiku 4.5 | $0.00109 | $0.00109 |
Grade A, and why
base 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.
How it starts
The opening of the file, as written. The whole thing — 99 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AI Agent Guidance
This file gives coding agents the repository-specific rules for Base. It is a navigation layer over the existing contributor docs, not a replacement for them.
Working Agreement
- Follow
CONTRIBUTING.mdfor workflow andSTANDARDS.mdfor code standards. - Keep Base focused as the local operating contract for deterministic readiness and handoff across independent Git repositories. Its durable loop is inventory -> prepare -> verify -> trust -> onboard -> hand off.
- Keep project-specific setup, service code, and application behavior in the owning project repository unless Base is explicitly the right shared layer.
- Adopt external agent workflow ideas only after translating them into Base-specific guidance. Do not vendor or require a third-party methodology when a smaller Base-native rule is enough.
- When the user explicitly says a session is design-only or asks for no code changes, stay in discussion mode and do not edit files.
- Surface unresolved product or architecture decisions instead of silently choosing defaults for broad changes.
GitHub Workflow
Before implementation, complete the pre-edit workflow gate and work from the issue-backed worktree. Use the GitHub Workflow for the canonical issue, repo-named Project, branch, worktree, pull request, and cleanup policy, including superseded pull requests. This file adds only the agent-specific repository guidance below.
Repository Release Contract
Repositories that publish versioned artifacts should opt into Base's release
standardization with basectl repo configure --release --repo <owner/name>.
This adds missing release metadata to base_manifest.yaml and a
docs/release-process.md guide without overwriting repository-specific release
content. Use basectl repo check --release to verify adoption, and read the
generated release guide before preparing a release.
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.
- 2d ago First seen · 99 lines · 1,086 tokens per session scan A 39ab211e1e7f
base AGENTS.md is an instructions file published in the GitHub repository basefoundry/base (119 stars, last pushed 2d ago), licensed Apache-2.0. It adds 1,086 tokens to every session, about $0.0054 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
lazy-starter-kit CLAUDE.md
Instructions for Heoooooon/lazy-starter-kit: Follow the safety rules in AGENTS.md. In particular, never execute recursive rm directly; use lazy-safe-rm for a strict descendant of the current Git workspace.
lazy-starter-kit AGENTS.md
Instructions for Heoooooon/lazy-starter-kit, a project described as: A signed GUI and one safe command to turn macOS, Linux, or Windows into a complete AI development environment — dry-run first, repeatable, and tested.
gentelella CLAUDE.md
Instructions for ColorlibHQ/gentelella, covering claude.md, what this is, commands, architecture and directory layout.
gentelella AGENTS.md
Instructions for ColorlibHQ/gentelella, covering agents.md, what this is, setup, architecture and directory layout.
gentelella copilot-instructions.md
Instructions for ColorlibHQ/gentelella, covering github copilot instructions — gentelella v4, hard rules, file layout, anti-patterns and commands.
maroto AGENTS.md
Instructions for johnfercher/maroto, covering agent instructions, skills, definition of done, contributing commands and tool dependencies.