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/joshuawangtw/rustbrowser/claude-mdgit clone --depth 1 https://github.com/JoshuaWangTW/RustBrowserWrote 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/joshuawangtw/rustbrowser/claude-md)<a href="https://agentmods.dev/instructions/joshuawangtw/rustbrowser/claude-md"><img src="https://agentmods.dev/badge/instructions/joshuawangtw/rustbrowser/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 | $0.00852 | $0.00852 |
| Opus 5 | $0.00426 | $0.00426 |
| Sonnet 5 | $0.00170 | $0.00170 |
| Haiku 4.5 | $0.00085 | $0.00085 |
Grade A, and why
RustBrowser 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.
What it actually says
RustBrowser 執行準則
每個 session 適用。RB = RustBrowser(僅為互動別名,不改 code 層命名)。互動慣例與 issue 記錄模板見 AGENTS.md。
模型分工 × Loop(固定準則)
- Fable 5 = 大腦(主迴圈):定可驗證的 goal、盤點排序、拆小輪、派工、每輪驗證、收斂記錄。判斷不外包。
- Opus = 派工/規劃層:稽核、架構、規劃、審查一律派 Opus agents(
architect/planner/critic/code-reviewer,model: opus)。 - Sonnet 5 = 執行層:規格已明確的實作派
executor(model: sonnet),每輪 1–3 筆、改動面小、可獨立驗證。 - Loop 直到 goal:依 loop-method——goal 未達成且仍能推進就繼續下一輪;停止條件 = goal 全數驗證通過,或連續 2 輪無法推進(此時回報卡點,不硬衝)。
- 例外:單檔小改動 Fable 直接做,不為小事開 agent。
宣告完成的唯一標準:跑完下方驗證流程且全綠 + 真實驗證留證據。絕不因「edit 成功」宣告完成。
Build 環境(Windows 受限網路,每個 shell 都要)
$env:Path = "$env:USERPROFILE\.cargo\bin;D:\aiproject\RustBrowser\.tools\nasm-2.16.03;$env:Path"
$env:CARGO_HTTP_CHECK_REVOKE = "false"
- TLS 走 rustls 雙根憑證(
rustls-tls-webpki-roots+rustls-tls-native-roots),不要改回 native-tls、也不要只留單一 rustls feature(會 UnknownIssuer)。 - 不引入需要 CMake 的依賴(aws-lc-rs 系禁止);能用現有依賴/stdlib 就不加新套件——輕量是產品定位。
驗證流程(= CI 同款,宣告完成前必跑)
cargo fmt --all -- --checkcargo clippy --all-targets -- -D warningscargo test(integration / session / eval / schema_freeze / benchmark 全綠)- 動過 fetch/render/MCP surface:用 release binary 對真實網站煙測一次,貼輸出片段當證據。
詳細步驟編碼於 .claude/skills/verify-rb/SKILL.md。
安全紅線(優先不要做)
- 不做 pixel-level click;不做完整 JS browser;不讓 LLM 未確認執行任意 POST;不把 RB 做成 Chrome clone。
- 公開 surface 受 semver 凍結(
docs/API.md+tests/schema_freeze.rs):additive 可以,breaking = major。
慣例
- Commit 中文描述
<type>: <描述>;feature branch → PR → merge(repo 既有模式)。 - RB 抓取問題記錄進
RB_FETCH_ISSUES.md;修好後在該 issue 下補記結案。 gh偶發 401:$env:GH_TOKEN = (gh auth token)後重試;merge 走 REST(gh api -X PUT .../pulls/N/merge)較穩。
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 First seen · 44 lines · 852 tokens per session scan A 69d13cda6b5d
RustBrowser CLAUDE.md is an instructions file published in the GitHub repository JoshuaWangTW/RustBrowser (10 stars, last pushed 15d ago), licensed MIT. It adds 852 tokens to every session, about $0.0043 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
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.
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 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.