Borrowing it
Nothing to install: this file belongs to Atlas-X-AI/shade-browser-security. 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/Atlas-X-AI/shade-browser-security/security-only/CLAUDE.mdgit clone --depth 1 https://github.com/Atlas-X-AI/shade-browser-securityWrote 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/atlas-x-ai/shade-browser-security/claude-md)<a href="https://agentmods.dev/instructions/atlas-x-ai/shade-browser-security/claude-md"><img src="https://agentmods.dev/badge/instructions/atlas-x-ai/shade-browser-security/claude-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/atlas-x-ai/shade-browser-security/claude-md"><img src="https://agentmods.dev/badge/instructions/atlas-x-ai/shade-browser-security/claude-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 | $0.06682 | $0.06682 |
| Opus 5 | $0.03341 | $0.03341 |
| Sonnet 5 | $0.01336 | $0.01336 |
| Haiku 4.5 | $0.00668 | $0.00668 |
Grade A, and why
shade-browser-security CLAUDE.md scanned grade A with 2 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
(`curl|sh` + PowerShell + PyPI) → browser device-code login (GitHub/Google/magic-link) → `onboard` Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
→ if dead: subprocess.run(["bash", LAUNCH_BRAVE_CDP, str(slot)], env=_display_env()) How it starts
The opening of the file, as written. The whole thing — 190 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.
What this is
A FastMCP server that exposes a Brave browser to Claude Code over CDP, across 20 isolated slots (ports 61000–61019). Each slot gets its own COPY of the user's real Brave profile (cookies + extensions + logins), so parallel agents do not collide on tabs or sessions. Distributed as a uv tool install-able Python package; registered with Claude Code via claude mcp add shade-browser-mcp --scope user -- shade-browser-mcp.
Two runtimes share the same tool surface:
- Local stdio (the default, what Claude Code invokes) — one operator, real on-disk Brave profiles, full tool set.
- Hosted multi-tenant HTTP (
serve --http, gated bySHADE_HOSTED) — streamable-HTTP transport with tenant API-key auth (auth.py), per-tenant slot quotas, fail-closed ephemeral profiles under/var/lib/shade, a hosted tool allowlist (_hosted_tool_allowed), egress blocklist, and the tokenized CDP auth proxy mandatory on every attach. Most security invariants below exist for this mode.
Common commands
Use uv (not pip) — uv.lock is checked in.
uv sync --extra dev # dev install + Playwright bindings (test deps are an extra, NOT --dev)
uv run --extra dev pytest # full unit suite (no Brave needed) — plain `uv run pytest` resolves an empty env (fake collection errors)
uv run --extra dev pytest tests/test_smoke.py # single file
uv run --extra dev pytest tests/test_smoke.py::test_package_imports # single test
uv run --extra dev pytest -k "not concurrent" # skip the concurrent-agents file
uv run --extra dev ruff check # lint
uv run --extra dev ruff check --fix # autofix
uv run shade-browser-mcp doctor # health check (Brave, profile, MCP registration, display env)
uv run shade-browser-mcp # start the stdio MCP server (what Claude Code invokes)
uv run shade-browser-mcp serve --http # streamable-HTTP transport (hosted multi-tenant mode)
uv run shade-browser-mcp launch-slot <slot> # launch one CDP slot for systemd/deploy use
uv run shade-browser-mcp dashboard # local Shade Fleet Console (aiohttp, 127.0.0.1 + bearer token)
uv run shade-browser-mcp tenant create|list|revoke # manage hosted HTTP tenant API keys
uv run shade-browser-mcp audit-verify # verify the tamper-evident hash chain of an event/audit log
uv run python -m shade_browser_mcp.cdp_proxy <slot> --require-token # per-slot CDP auth proxy (manual)
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 · 190 lines · 6,682 tokens per session scan A 91dfa8dfa4f7
shade-browser-security CLAUDE.md is an instructions file published in the GitHub repository Atlas-X-AI/shade-browser-security (0 stars, last pushed 1mo ago), licensed MIT. It adds 6,682 tokens to every session, about $0.0334 per session on Opus 5. A static security scan graded it A with 2 findings (makes network calls, runs shell commands). 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.