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/sublimecoder/aios/claude-mdgit clone --depth 1 https://github.com/sublimecoder/aiosWhat 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 | $0.01610 | $0.01610 |
| Opus 5 | $0.00805 | $0.00805 |
| Sonnet 5 | $0.00322 | $0.00322 |
| Haiku 4.5 | $0.00161 | $0.00161 |
Grade A, and why
aios 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.
How it starts
The opening of the file, as written. The whole thing — 64 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE — AI OS entry point
This vault is an AI Operating System (plain markdown, no plugins, portable across AI tools). Everything the AI needs lives in AIOS/.
New vault? Run
/aios-bootstrap(or pasteBOOTSTRAP.mdinto your agent) once. It interviews you, fills in the<placeholders>below, and rewrites this file.
Start here, every session
- Read
AIOS/Systems/layers.tsv— the scopes this vault partitions knowledge into. A single-scope vault is normal; if there's one row, skip the ceremony. - Read the me file for the scope this task belongs to:
<scope>/me.md. - Read
AIOS/Maps/Vault Map.md— how to navigate and where new notes go. - Read
AIOS/Maps/Skill Map.md— every skill/system and its trigger. Skills live inAIOS/Skills/, systems inAIOS/Systems/.
Scopes
A scope is a bucket of context — a job, a side project, a client, a life area. Its job is routing: it keeps one project's accumulated memory out of an unrelated one, and tells the AI which me.md file to read.
- The scope is the filesystem: each scope is one top-level directory named by its slug —
<scope>/me.md,<scope>/notes/,<scope>/content/,<scope>/projects/,<scope>/sources/. A path's first segment names its scope; that prefix is what the write guard, ingest, and leak audits key on. - Every content note also carries a
layer:frontmatter field naming its scope. - Ingest, query, and lint each work one scope at a time, so a synthesis note never blends two unrelated contexts. This is about keeping knowledge coherent, not about secrecy.
Strict separation (optional — off by default)
Most vaults do not need this. Skip the whole section unless it applies to you.
Some people keep scopes that must never appear in the same file — a pen name that can't be linked to a legal name, employer-proprietary code that can't reach a public portfolio or a job application, one client's internals that can't surface in another's repo. If that's you, enable the wall by filling in the tokens column of AIOS/Systems/layers.tsv.
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.
- yesterday First seen · 64 lines · 1,610 tokens per session scan A b71da4a97559
aios CLAUDE.md is an instructions file published in the GitHub repository sublimecoder/aios (2 stars, last pushed 1mo ago), licensed MIT. It adds 1,610 tokens to every session, about $0.0080 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 2026-08-31.
Other instructions, from other repositories
personalos-boilerplate AGENTS.md
Instructions for vincentmumme/personalos-boilerplate, covering personalos boilerplate, einstieg mit einem nutzer, verbindliche grenzen, maintainer-arbeit and sprache.
brain-starter CLAUDE.md
Instructions for michaeljauk/brain-starter, covering brain repo - claude instructions, purpose, not set up yet?, file placement rules and directory guide.
commonplace CLAUDE.md
Instructions for briannaworkman/commonplace: This repository holds two vaults. Nothing should be ingested or edited here at the root.
markdown-agent-vault-ru AGENTS.md
Instructions for dzhokhov/markdown-agent-vault-ru, covering agents.md — правила агента для файлового хранилища, главный принцип, карта папок, загрузка контекста and недосказанные запросы.
memory-forest AGENTS.md
AGENTS.md instructions for hyungchulc/memory-forest, covering repository contract, scope, invariants and change discipline.
obsidian-mind CLAUDE.md
Claude Code instructions for breferrari/obsidian-mind, covering obsidian mind, skills & capabilities, custom slash commands, vault structure and obsidian cli.