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/shck-dev/notion-mcp/claude-mdgit clone --depth 1 https://github.com/shck-dev/notion-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/shck-dev/notion-mcp/claude-md)<a href="https://agentmods.dev/instructions/shck-dev/notion-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/shck-dev/notion-mcp/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 | $0.02310 | $0.02310 |
| Opus 5 | $0.01155 | $0.01155 |
| Sonnet 5 | $0.00462 | $0.00462 |
| Haiku 4.5 | $0.00231 | $0.00231 |
Grade A, and why
notion-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 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.
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 — 111 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Notion MCP Server
MCP server for Notion using internal API (cookie auth, token_v2). No workspace admin or OAuth needed.
Package
- npm:
@shck-dev/notion-mcp(currently 0.5.2) - GitHub:
shck-dev/notion-mcp - Runtime: ships a bundled, node-compatible
dist/server.js(#!/usr/bin/env node), sonpx @shck-dev/notion-mcpworks without Bun. Built withbun build(bun run build); dev/test still use Bun. - License: MIT
Architecture
src/
├── server.ts # Thin stdio entry; `init` argv → CLI wizard
├── mcp-handler.ts # MCP protocol: TOOLS registry + JSON-RPC routing (testable, no transport)
├── notion-client.ts # loadConfig() (env → file), notionPost(), parsePageId()
├── config-store.ts # ~/.notion-mcp/config.json read/write (0600); NOTION_MCP_CONFIG_DIR override
├── cli-init.ts # `notion-mcp init` interactive wizard (reads cURL from stdin)
├── guide.ts # SETUP_GUIDE text (shared: CLI, prompt, resource)
├── prompts.ts # notion_setup, notion_search_export
├── resources.ts # notion://guide (static), notion://recent (dynamic)
├── transport.ts # stdio JSON-RPC newline-delimited transport
├── notion-files.ts # uploadImageFile(), uploadImagesForBlocks(), buildImagePatchOps(), resolveImages()
├── types.ts # NotionConfig, NotionBlock interfaces
├── tools/
│ ├── init.ts # parseCurl(), resolveSpaces(), notionInit() — the notion_init tool
│ ├── search.ts # notion_search
│ ├── export.ts # notion_export_page (resolves images; accepts image_dir for local download)
│ ├── import.ts # notion_import_page(_from_file), notion_append_to_page(_from_file), buildCreateOps()
│ ├── create-page.ts # notion_create_page, notion_create_page_from_file
│ ├── images.ts # notion_add_image (local upload or external URL reference)
│ └── comments.ts # notion_list_comments, notion_add_comment (anchor_text), notion_reply_comment
└── markdown/
├── to-notion.ts # richText() + markdownToNotionBlocks()
└── from-notion.ts # richTextToMarkdown() + blocksToMarkdown() (child page → sub-page link)
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 · 111 lines · 2,310 tokens per session scan A d9669a8550ae
notion-mcp CLAUDE.md is an instructions file published in the GitHub repository shck-dev/notion-mcp (2 stars, last pushed 2mo ago), licensed MIT. It adds 2,310 tokens to every session, about $0.0115 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
inkdex AGENTS.md
AGENTS.md instructions for antonlunden/inkdex, covering agents.md, general coding guidelines and development.
lightpaper CLAUDE.md
Claude Code instructions for lightpaperorg/lightpaper, covering lightpaper.org — claude code instructions, architecture, key files, running locally and running tests.
lightpaper AGENTS.md
AGENTS.md instructions for lightpaperorg/lightpaper, covering agents.md — lightpaper.org, what is lightpaper.org?, quick start, mcp server (recommended for agents) and direct api.
precis-mcp AGENTS.md
AGENTS.md instructions for retospect/precis-mcp, covering precis-mcp — agent context, identity, stack / toolkit, repository shape and workflow — plan first, always.
precis-mcp CLAUDE.md
Claude Code instructions for retospect/precis-mcp, covering claude code — project brief, ship workflow, orientation, conventions that bite (irreversible, or reddens the ship) and hook/gate-enforced — one-liners, detail on demand.
markitdown-mcp-server AGENTS.md
AGENTS.md instructions for RECTOR-LABS/markitdown-mcp-server, covering apify actors development guide, what are apify actors?, core concepts, do and don't.