baseline CLAUDE.md

baseline CLAUDE.md is an instructions file for coding agents from friedbotstudio/baseline. It costs 6,944 tokens per session, scanned A, original, Apache-2.0.

A set of project instructions that defines how Claude Code should operate in a specific codebase. It establishes rules, priorities, required startup checks, and workflow steps.

In plain words
What is it for?
Use it to enforce repository rules, control which instructions take precedence, and guide coding workflows.
Why use it?
It gives the coding agent consistent project-specific guidance instead of relying only on general defaults.

Instructions file

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/friedbotstudio/baseline/claude-md
Clone the repo
git clone --depth 1 https://github.com/friedbotstudio/baseline

Wrote 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.

agentmods badge for baseline CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/friedbotstudio/baseline/claude-md.svg)](https://agentmods.dev/instructions/friedbotstudio/baseline/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/friedbotstudio/baseline/claude-md"><img src="https://agentmods.dev/badge/instructions/friedbotstudio/baseline/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 6,944 This file is loaded in full into every session.
When invoked 6,944 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 $0.06944 $0.06944
Opus 5 $0.03472 $0.03472
Sonnet 5 $0.01389 $0.01389
Haiku 4.5 $0.00694 $0.00694

Measured yesterday against content hash 50fddf41480e, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

baseline CLAUDE.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 yesterday.

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.

CLAUDE.md · 279 lines

How it starts

The opening of the file, as written. The whole thing — 279 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Claude Code Baseline — In-Session Constitution

This file binds Claude's behavior — Claude Code, in this codebase, in this session. Every rule below is mandatory unless waived by an explicit exceptions entry in .claude/state/workflow.json (set only by /triage).

Genesis prompt. docs/init/seed.md is the governing specification of this baseline. When it and this constitution conflict, seed.md governs and you SHALL stop and surface the drift before acting. When this constitution and the implementation conflict, this constitution governs and the implementation SHALL be corrected.

Enforcement. The 27 hooks in .claude/hooks/ are the enforcement layer, mapped to their Articles in Article VIII.


Article I — Authority and precedence

  1. Genesisdocs/init/seed.md is the source of truth for the baseline's shape, components, and rebuild protocol.
  2. ConstitutionCLAUDE.md is the source of truth for Claude's in-session behavior.
  3. Implementation — hooks, skills, commands, subagent, MCP servers, and config actuate and enforce (1) and (2).
  4. Order of precedenceseed.md > CLAUDE.md > implementation. Lower binds higher only via amendment in seed.md, which then propagates to this file, then to disk.
  5. Project amendments — Article XI reserves space for project-owner amendments. They bind alongside the baseline Articles but SHALL NOT contradict them.
  6. Size cap (this file)CLAUDE.md SHALL NOT exceed 40,000 characters and carries binding rules only; history, narration, and reference appendices live in the annex. audit-baseline enforces the cap, which also binds the byte-equal mirror src/CLAUDE.template.md. The governance suite pins a tighter advisory target (seed.md §14).

Article II — Architectural principle

Decisions live in main context. Subagents only execute pre-decided recipes in parallel or in the background.

The baseline ships exactly one subagent: swarm-worker. Its sole sanctioned use is Skill(scenario) then Skill(implement) against a fully-specified recipe in an isolated git worktree during /swarm-dispatch. It SHALL NOT make design choices, pick abstractions, or expand scope, and SHALL NOT be invoked outside /swarm-dispatch.

Read the full file on GitHub · 279 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. yesterday First seen · 279 lines · 6,944 tokens per session scan A 50fddf41480e

Subscribe to this mod's changes

baseline CLAUDE.md is an instructions file published in the GitHub repository friedbotstudio/baseline (14 stars, last pushed yesterday), licensed Apache-2.0. It adds 6,944 tokens to every session, about $0.0347 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 2026-09-03.