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/addyosmani/factory/claude-mdgit clone --depth 1 https://github.com/addyosmani/factoryWrote 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/addyosmani/factory/claude-md)<a href="https://agentmods.dev/instructions/addyosmani/factory/claude-md"><img src="https://agentmods.dev/badge/instructions/addyosmani/factory/claude-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 | $0.00414 | $0.00414 |
| Opus 5 | $0.00207 | $0.00207 |
| Sonnet 5 | $0.00083 | $0.00083 |
| Haiku 4.5 | $0.00041 | $0.00041 |
Grade B, and why
factory CLAUDE.md scanned grade B with 1 finding 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 4d 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.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
jq empty template/.claude/settings.json template/.codex/hooks.json How it starts
The opening of the file, as written. The whole thing — 53 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Factory reference development
This repository builds and documents a reference software factory. It is the distribution
repository, not an installed factory. Files under template/ are copied into another
repository by install.sh.
Read before changing behavior
README.mdis the public map and product boundary.GETTING_STARTED.mdis the desktop-first tutorial.ARCHITECTURE.mdexplains the design decisions.LIMITS.mdrecords product constraints that may change over time.template/docs/factory/CONTRACT.mdis the harness-neutral installed policy.
Preserve the distinction between this repository's contributor instructions and the
template/CLAUDE.md instructions installed in user projects.
Validation
Run the complete local suite before committing:
bash tests/run.sh
For a focused syntax check:
bash -n install.sh template/.claude/hooks/*.sh template/.claude/scripts/*.sh \
template/.factory/scripts/*.sh tests/*.sh
jq empty template/.claude/settings.json template/.codex/hooks.json
The tests use disposable repositories. Do not point them at a real project.
Change rules
- Keep
install.shidempotent and refuse to overwrite user files. - Keep Claude Code as the canonical workflow definition. Codex adapters should remain thin and point to the same contract and workflow files.
- Treat the charter, required gates, merge boundary, and independent verification rule as protected behavior. A convenience change must not silently weaken them.
- Update the tutorial when a user-visible install step changes.
- Date product-specific claims in
LIMITS.mdand verify them against official documentation before changing them. - Use small commits whose message names the behavior or documentation boundary changed.
Writing
Write for a programmer trying this on a real repository. Prefer exact paths, commands, expected output, and failure behavior over broad claims. State when Claude and Codex differ instead of smoothing over the difference.
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.
- 4d ago First seen · 53 lines · 414 tokens per session scan B 6c1fea747cd1
factory CLAUDE.md is an instructions file published in the GitHub repository addyosmani/factory (172 stars, last pushed 13d ago), licensed MIT. It adds 414 tokens to every session, about $0.0021 per session on Opus 5. A static security scan graded it B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
software-factory-starter AGENTS.md
AGENTS.md instructions for artreimus/software-factory-starter, covering agents.md, core rules, artifact boundaries, skill routing and engineering standards.
ox AGENTS.md
AGENTS.md instructions for sageox/ox, covering human attention is the scarcest resource, what is ox?, quick start, terminology and required reviews.
fluent AGENTS.md
AGENTS.md instructions for mrinalwadhwa/fluent, covering instructions for coding agents, fluent workflow, commit messages, subject line and body.
turbo-flow CLAUDE.md
Claude Code instructions for marcuspat/turbo-flow, covering claude.md — turbo flow, what this repo is, key files and conventions.
alignfirst AGENTS.md
AGENTS.md instructions for paleo/alignfirst, covering repository guidelines, tooling, packages, docmap - seek documentation and workspaces.
agent-skills AGENTS.md
AGENTS.md instructions for a-tokyo/agent-skills, covering agents.md, project overview, repo structure (monorepo), setup & commands and install / use a skill (consumer side).