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/54334-bit/anki_mcp_server/claude-mdgit clone --depth 1 https://github.com/54334-bit/anki_mcp_serverWrote 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/54334-bit/anki_mcp_server/claude-md)<a href="https://agentmods.dev/instructions/54334-bit/anki_mcp_server/claude-md"><img src="https://agentmods.dev/badge/instructions/54334-bit/anki_mcp_server/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 5.1 | $0.04236 | $0.04236 |
| Opus 5 | $0.02118 | $0.02118 |
| Sonnet 5 | $0.00847 | $0.00847 |
| Haiku 4.5 | $0.00424 | $0.00424 |
Grade A, and why
anki_mcp_server 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 5d 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 — 304 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
Project Overview
MCP server enabling AI assistants to interact with Anki via AnkiConnect. Built with NestJS and @rekog/mcp-nest.
- Package:
@ankimcp/anki-mcp-server(npm) - License: MIT
- Status: Beta (0.x.x) - breaking changes allowed
- User-facing docs:
README.mdcovers installation, client setup (Claude Desktop/MCPB, HTTP, tunnel), and the full tool catalog — consult it for anything user-facing rather than reconstructing it here.
Quick Reference
# Build & Run
npm run build # Build → dist/ (all three entry points)
npm run start:dev:stdio # STDIO mode with watch
npm run start:dev:http # HTTP mode with watch
npm run start:dev:tunnel # Tunnel mode with watch + debug
# Testing
npm test # All tests
npm test -- path/to/file.spec.ts # Single test file
npm run test:tools # Tool unit tests only
npm run test:workflows # Multi-tool workflow scenarios (mocked)
npm run test:cov # With coverage (70% threshold)
npm run e2e:full:local # One-shot E2E: up → test → down
# Quality
npm run lint && npm run type-check # Pre-push checks (also runs via Husky)
# Debugging
npm run inspector:stdio # MCP Inspector UI for testing tools
npm run inspector:stdio:debug # With debugger on port 9229
npm run inspector:http # Inspector against HTTP transport
Architecture
Entry Points
Three entry points compiled in a single build:
| Mode | Entry | Use Case | Logging |
|---|---|---|---|
| STDIO | dist/main-stdio.js |
Claude Desktop, MCP clients | stderr |
| HTTP | dist/main-http.js |
Web-based AI (ChatGPT, claude.ai) | stdout |
| Tunnel | dist/main-tunnel.js |
Remote access via WebSocket tunnel | stderr |
Core Files
src/
├── main-stdio.ts # STDIO bootstrap: NestFactory.createApplicationContext()
├── main-http.ts # HTTP bootstrap: NestFactory.create() + guards
├── main-tunnel.ts # Tunnel bootstrap: auth commands + WebSocket tunnel
├── app.module.ts # Root module with forStdio()/forHttp()/forTunnel() factories
├── bootstrap.ts # Shared logger setup (pino → NestJS LoggerService)
├── version.ts # Package version constant (read from package.json at build)
├── cli/ # CLI layer (Commander parsing + user-facing output)
│ ├── args.ts # Commander entrypoint: option parsing, subcommand dispatch
│ ├── cli-output.ts # User-facing print helpers (cli.success/error/info/box) — separate from args.ts so non-CLI code can use it without pulling in Commander
│ ├── spinner.ts # Terminal spinner for long-running CLI operations
│ └── index.ts # Barrel re-export for the cli module
├── app-config.service.ts # IAnkiConfig implementation (reads from validated AppConfig)
├── config/ # Zod-validated config system (schema, factory, APP_CONFIG token)
├── services/ngrok.service.ts # Optional ngrok subprocess for HTTP-mode public tunneling
├── http/guards/ # HTTP-only guards (Origin allowlist)
├── tunnel/ # Tunnel mode: WebSocket client, OAuth device flow, credentials
│ ├── tunnel.client.ts # WebSocket client for tunnel server
│ ├── tunnel-mcp.service.ts # Bridges MCP ↔ tunnel via InMemoryTransport
│ ├── in-memory.transport.ts # MCP transport that connects to TunnelClient in-process
│ ├── tunnel.protocol.ts # WS message type definitions (request/response/ping/error)
│ ├── device-flow.service.ts # OAuth device flow authentication
│ ├── credentials.service.ts # Persistent credential storage
│ └── commands/ # CLI command handlers (login, logout, tunnel)
└── mcp/
├── clients/anki-connect.client.ts # HTTP client using ky (retries, error handling, read-only guard)
├── config/anki-config.interface.ts # ANKI_CONFIG injection token + IAnkiConfig interface
├── types/anki.types.ts # Shared Anki types (cards, notes, ratings)
├── utils/ # Shared utilities (anki.utils, markdown.utils, stats.utils)
├── primitives/essential/ # Core tools, prompts, resources
└── primitives/gui/ # GUI-specific tools (require user approval)
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.
- 5d ago First seen · 304 lines · 4,236 tokens per session scan A 991d135a7f48
anki_mcp_server CLAUDE.md is an instructions file published in the GitHub repository 54334-bit/anki_mcp_server (0 stars, last pushed 5d ago), licensed MIT. It adds 4,236 tokens to every session, about $0.0212 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
open-dictionary AGENTS.md
Instructions for ahpxex/open-dictionary, covering open dictionary rewrite charter, product framing, core workflow, technical framework and 1. raw ingestion layer.
ab900 CLAUDE.md
Instructions for timothywarner-org/ab900, covering claude.md, repository purpose, architecture, commands and 2. non-ascii punctuation -- must return zero.
canvas-lms-mcp AGENTS.md
AGENTS.md instructions for bruchris/canvas-lms-mcp, covering agents.md — canvas lms mcp server, quick start, run with npx (no install needed), or install globally and architecture.
obsidian-university-workflow CLAUDE.md
Instructions for ABO896/obsidian-university-workflow, covering obsidian university workflow, project structure, rules for working on this project, templater api — always read the docs first and config alignment.
agentic-ai-engineering-course AGENTS.md
AGENTS.md instructions for towardsai/agentic-ai-engineering-course, covering the what and course map (lessons ↔ code projects).
anki-mcp-server CLAUDE.md
Claude Code instructions for nailuoGG/anki-mcp-server, covering claude.md, repository overview, development commands, build & development and install dependencies.