Borrowing it
Nothing to install: this file belongs to ManudotaORG/artist-mcp. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/ManudotaORG/artist-mcp/main/CLAUDE.mdgit clone --depth 1 https://github.com/ManudotaORG/artist-mcpWrote 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/manudotaorg/artist-mcp/claude-md)<a href="https://agentmods.dev/instructions/manudotaorg/artist-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/manudotaorg/artist-mcp/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/manudotaorg/artist-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/manudotaorg/artist-mcp/claude-md.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.03439 | $0.03439 |
| Opus 5 | $0.01720 | $0.01720 |
| Sonnet 5 | $0.00688 | $0.00688 |
| Haiku 4.5 | $0.00344 | $0.00344 |
Grade A, and why
artist-mcp 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 6d 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 — 231 lines — stays where its author put it; the contents beside it link to each section on GitHub.
artist-mcp
An npm-published MCP server that reads a user's OneNote notes, signing in to Microsoft and Google on the user's own machine, plus a small web app for sign-in and the install instructions.
Scope, what shipped, and the open gaps are docs/scope.md. Read it before starting work. It is the source of truth for scope and for what is actually done — a ticked box means done and verified, and blocked items say why. Update it as you go rather than at the end.
Scope is deliberately narrow. The workflow layer is defined in
apps/mcp/agent-pack: one OneNote page is one working unit, Markdown roles and
project types are loaded at runtime, and every result stays in chat.
If you find yourself adding writes, sends, or synchronization, stop. That rule stands, with four exceptions, and none of them is a precedent.
An install granted --allow-writes calendar-create may create a single Google
Calendar event, previewed and confirmed, and one granted calendar-delete may
remove an event that this tool itself created, identified by the artist
prefix on its id. An install holding both may also reschedule one, which is
those two writes in one confirmed step — a create then a delete, never a
PATCH, because the event id is a hash of the event's own contents. An event
the musician made is unreachable, and that prefix check is the only thing
making delete safe to offer. Read
docs/decisions/0001-opt-in-calendar-writes.md
before touching that path — it says what was decided, what it cost, and what
would reverse it.
An install granted onenote-create may create a new OneNote page, previewed
and confirmed. That capability alone cannot edit or delete any page, including
one it created, and that is not a rule this repository keeps — the scope is
Notes.Create, which cannot express an edit or a delete, verified as a 403 on
both against a page the token had just created itself. This is the inverse of
the calendar situation, where no insert-only scope exists and the boundary had
to become our code. Read
docs/decisions/0003-onenote-writes.md
before touching that path.
Editing is a separate grant, and it shipped. Notes.ReadWrite would hand
back edit and delete over every page and put the boundary in our code again —
but it was not the only door. An install granted onenote-edit holds
Notes.ReadWrite.CreatedByApp, which Microsoft enforces against this app's
own pages, verified as 401 on a page the musician wrote, and under it an
element-level replace is both surgical and recoverable from a pre-image captured
before every write — no capture, no write.
docs/decisions/0004-onenote-page-maintenance.md
was built and verified end to end on 2026-08-31, and that live run is what found
its last two defects; nothing else did.
0006 extends it to tables,
which is where a filled-in page actually keeps its content: OneNote supports no
update to a row or a cell, so the unit is the whole table, written as markup and
previewed as rows. That widens what one confirmed change can destroy, and it
amends policy:patch's smallest-fragment rule rather than quietly breaking it —
read it before touching either. Deleting stays out.
Message sending and synchronization remain out, and the reasoning is in the records rather than restated here.
Sources are read-only apart from those grants, and deliberately few. OneNote holds the working unit; Gmail and Google Calendar are supporting evidence only — they corroborate or fill gaps in a page and are never themselves a working unit. That asymmetry is the whole reason further sources did not dissolve the one-page-one-unit rule that every role and playbook depends on. Each new source means re-deciding it, not repeating it: Google Tasks was considered and left out, because a task list is a rival system of record for the work itself rather than evidence about it.
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.
- 6d ago Changed · +44 lines · +704 tokens per session de5952b23f1e
- 10d ago First seen · 187 lines · 2,735 tokens per session scan A 8ccf2e65acac
artist-mcp CLAUDE.md is an instructions file published in the GitHub repository ManudotaORG/artist-mcp (1 stars, last pushed 3d ago), licensed MIT. It adds 3,439 tokens to every session, about $0.0172 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
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
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).
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.