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 rules/yuz9610/poe2overlay/architecturegit clone --depth 1 https://github.com/yuz9610/POE2OverlayWrote 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/rules/yuz9610/poe2overlay/architecture)<a href="https://agentmods.dev/rules/yuz9610/poe2overlay/architecture"><img src="https://agentmods.dev/badge/rules/yuz9610/poe2overlay/architecture.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 | $0.00850 | $0.00850 |
| Opus 5 | $0.00425 | $0.00425 |
| Sonnet 5 | $0.00170 | $0.00170 |
| Haiku 4.5 | $0.00085 | $0.00085 |
Grade A, and why
architecture 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 — 51 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Architecture (PoE2 Area Overlay)
Read docs/README.md for docs ownership and docs/ARCHITECTURE.md for full technical context. Summary:
Workflow canonical rules: docs/PROJECT_WORKFLOW.md. Gate summary: .cursor/rules/workflow.mdc. Prompts: docs/AI_AGENT_PROMPTS.md. Human cheat sheet: docs/AI_AGENT_SETUP.md.
Workflow short commands: follow matching step in AI_AGENT_PROMPTS.md; first response = proposal only unless skip phrase in same message (see workflow.mdc).
Docs architecture: every docs/**/*.md should begin with 文檔目的 / 不負責. Keep one source of truth per knowledge type: architecture/IPC → ARCHITECTURE.md; user-facing behavior → USER_GUIDE.md; build schema → BUILD_JSON.md; build generation workflow → CREATE_BUILD.md; workflow rules → PROJECT_WORKFLOW.md; prompts → AI_AGENT_PROMPTS.md; agent maintenance → AI_AGENT_SETUP.md; current state → BRAINSTORM.md / SPRINTS.md / docs/sprints/.
Stack
Tauri 2 + Rust (src-tauri/) · React 19 + TS + Vite 6 (src/) · JSON data (data/). Local-only; no HTTP/shell in Rust.
Where things live
- IPC:
src/api/tauri.ts↔ commands insrc-tauri/src/lib.rs - App state:
src/app/useOverlayData.ts(single source for settings, area, build, profiles, reminders) - Log pipeline:
watcher.rs→log_parser.rs→ emitarea-changed/character-changed - Hints:
area_hints.rs+data/areas.json(+ optionaldata/build/override) - Types:
src/types.ts - Build JSON:
docs/BUILD_JSON.md（schema）、docs/CREATE_BUILD.md（AI 生成 BD） - Project workflow/state: rules in
docs/PROJECT_WORKFLOW.md; prompts indocs/AI_AGENT_PROMPTS.md; current state indocs/SPRINTS.md,docs/BRAINSTORM.md,docs/sprints/
Agent rules
- Encoding: UTF-8 no BOM, LF — see
docs/ENCODING.mdandencoding.mdc - New IPC: update Rust command +
tauri.ts+types.tsif needed - Avoid duplicate
listenon same Tauri event; pass static config via props fromuseOverlayData - Background locks: use
sync_util::{mutex_lock,rw_read,rw_write}in Rust worker threads - Chinese area names in JSON must match
Client.txtbytes exactly - Docs sync: feature/behavior changes must update matching docs — see
docs-sync.mdcand ARCHITECTURE § 文檔同步 - Workflow sync: task/sprint/commit flow — follow
docs/PROJECT_WORKFLOW.mdand.cursor/rules/workflow.mdc(proposal before writeback; Step 4 two phases) - Docs ownership: do not duplicate long rules across docs; update the SOT named in
docs/README.mdand keep this rule / AGENTS as summaries only
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 · 51 lines · 850 tokens per session scan A 41b0515c1821
architecture is a cursor rule published in the GitHub repository yuz9610/POE2Overlay (9 stars, last pushed 3mo ago), licensed MIT. It adds 850 tokens to every session, about $0.0042 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 cursor rules, from other repositories
development-conventions
Development conventions, code style, and best practices for the Kaneo project.
core
Core STT/TTS abstraction layer documentation.
new-component
Workflow for creating new React components.
middleware-auth
Next.js middleware auth — prevents auth logic being placed in middleware instead of route handlers.
react-basics-state
React 状态管理 - Context API、useReducer、AuthContext 模式.
project-structure
Project folder layout for Vite and Next.js variants.