Borrowing it
Nothing to install: this file belongs to luiz-santos-it/anki-mcp-addon. 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/luiz-santos-it/anki-mcp-addon/master/AGENTS.mdgit clone --depth 1 https://github.com/luiz-santos-it/anki-mcp-addonWrote 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/luiz-santos-it/anki-mcp-addon/agents-md)<a href="https://agentmods.dev/instructions/luiz-santos-it/anki-mcp-addon/agents-md"><img src="https://agentmods.dev/badge/instructions/luiz-santos-it/anki-mcp-addon/agents-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/luiz-santos-it/anki-mcp-addon/agents-md"><img src="https://agentmods.dev/badge/instructions/luiz-santos-it/anki-mcp-addon/agents-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 | $0.01439 | $0.01439 |
| Opus 5 | $0.00720 | $0.00720 |
| Sonnet 5 | $0.00288 | $0.00288 |
| Haiku 4.5 | $0.00144 | $0.00144 |
Grade A, and why
anki-mcp-addon AGENTS.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 8d 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.
Copies of this mod
1 near-identical copy found in the catalogue:
- anki-mcp-addon CLAUDE.md — 100% identical, 0 lines differ
How it starts
The opening of the file, as written. The whole thing — 71 lines — stays where its author put it; the contents beside it link to each section on GitHub.
anki-mcp-addon
Python Anki add-on that exposes Anki as MCP tools via an in-process SSE server. Zero external deps; calls Anki's internal Python APIs directly.
Commands
pytest # run all tests (no Anki required)
pytest -x # stop on first failure
pytest tests/test_notes.py # single file
Architecture
__init__.py ← Anki entry point: registers tools, starts server, wires config UI/hooks
config_ui.py ← Qt settings dialog: Server tab (port/status/restart) + Client Setup tab
(one-click Claude Code/Desktop connection helpers + .mcpb export)
client_setup.py ← pure port→command/JSON/URL string builders + .mcpb copy; no Qt/Anki
imports, tested directly without mocking aqt.qt
protocol.py ← JSON-RPC 2.0 + MCP dispatch; tool registry (_tools dict)
server.py ← ThreadingHTTPServer; GET /sse (SSE stream), POST /messages
tools/
notes.py ← 7 tools: create/search/update/delete/move notes + tag management
decks.py ← 3 tools: create/list/delete decks
study.py ← 5 tools: get_due_cards, submit_answer, reschedule, suspend, forget
insights.py ← 1 tool: retention stats via direct SQL (mw.col.db)
sync.py ← 1 tool: trigger AnkiWeb sync
search.py ← shared deck-name quoting for Anki search syntax
tests/
conftest.py ← fakes aqt/anki via sys.modules before any imports
desktop-extension/ ← Claude Desktop .mcpb SOURCE (manifest.json + server/index.js, runs a
bundled mcp-remote IN-PROCESS via dynamic import); packed via `mcpb pack`
desktop-extension.mcpb ← pre-built, tracked binary — the actual file client_setup.py's
export_mcpb() copies; rebuild+commit whenever the source above changes
scripts/build_ankiaddon.py ← builds the AnkiWeb .ankiaddon zip; must include client_setup.py
and desktop-extension.mcpb alongside the other runtime files
Key conventions
- Thread safety: tool functions call
mw.coldirectly;server.pyroutes all tool execution to the Qt main thread viamw.taskman.run_on_main()+concurrent.futures.Future. - No auth, Origin check instead:
server.pyrejects any request carrying anOriginheader — only browsers send one, no legitimate MCP client does. This is the CSRF defense (there's no other auth), don't add CORS headers back. - Server lifecycle:
server.stop()must call bothshutdown()andserver_close(), or the port stays bound and a same-port restart fails. - Benign connection errors:
_MCPServer.handle_error()and theexcept _BENIGN_CONNECTION_ERRORSin_handle_sseboth swallowBrokenPipeError/ConnectionResetError/ConnectionAbortedError— a client disconnecting mid-transport-negotiation is routine (confirmed with real Claude Desktop/mcp-remote traffic), not a bug worth a traceback in Anki's console. - Scheduler compat:
study.pychecksmw.col.v3_scheduler()and branches between v3answer_card(CardAnswer(...))and legacyanswerCard(card, ease). - v3 state cache:
_queued_statesinstudy.pymapscard_id → QueuedCard. Populated byget_due_cards(merged, not cleared, so concurrent decks don't clobber each other), consumed bysubmit_answer, and dropped viareset_queue()on profile open so stale state from a previous collection can't leak in. - Deck-name escaping: build search queries with
tools/search.py:quote_deck(), never raw f-string interpolation — deck names can contain". - SQL in insights:
mw.col.db.all(sql, *args)— positional params, not list. - No external deps: stdlib only (
http.server,threading,queue,json,uuid,concurrent.futures).
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.
- 8d ago First seen · 71 lines · 1,439 tokens per session scan A 293f26102681
anki-mcp-addon AGENTS.md is an instructions file published in the GitHub repository luiz-santos-it/anki-mcp-addon (1 stars, last pushed 1mo ago), licensed MIT. It adds 1,439 tokens to every session, about $0.0072 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).
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.
deepseek-harness AGENTS.md
AGENTS.md instructions for deepseek-ai/deepseek-harness, covering agents.md, pre-stable apis and released session data, repository layout, commands and host sandbox failures.