Borrowing it
Nothing to install: this file belongs to appclawhq/appium-mcp-auth. 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/appclawhq/appium-mcp-auth/main/CLAUDE.mdgit clone --depth 1 https://github.com/appclawhq/appium-mcp-authWrote 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/appclawhq/appium-mcp-auth/claude-md)<a href="https://agentmods.dev/instructions/appclawhq/appium-mcp-auth/claude-md"><img src="https://agentmods.dev/badge/instructions/appclawhq/appium-mcp-auth/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/appclawhq/appium-mcp-auth/claude-md"><img src="https://agentmods.dev/badge/instructions/appclawhq/appium-mcp-auth/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 | $0.01802 | $0.01802 |
| Opus 5 | $0.00901 | $0.00901 |
| Sonnet 5 | $0.00360 | $0.00360 |
| Haiku 4.5 | $0.00180 | $0.00180 |
Grade A, and why
appium-mcp-auth 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 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.
How it starts
The opening of the file, as written. The whole thing — 123 lines — stays where its author put it; the contents beside it link to each section on GitHub.
appium-mcp-auth — project guide
Authentication & authorization for appium-mcp when hosted over SSE / HTTP Stream, built entirely as an appium-mcp plugin. There are no changes to appium-mcp core — this is a hard project constraint (see below).
Golden rule: never modify appium-mcp core
All auth logic lives in this package as a plugin. Do not edit
../appium-mcp/src/* (e.g. create-server.ts, plugin.ts) to add auth. A
plugin cannot read HTTP headers, so the credential is passed as a tool
argument (authToken), and both authentication and authorization happen in
the plugin's beforeCall hook. If a change seems to require touching core,
stop and surface it — don't fork core.
Architecture
One process = the full appium-mcp SSE server with this plugin composed in.
Users run this package's CLI instead of appium-mcp's bare CLI.
tool call ──► AppiumAuthPlugin.beforeCall(ctx)
1. authenticate ctx.args.authToken → Identity (authenticator.ts)
2. rate-limit per subject (rate-limiter.ts)
3. authorize tool vs Identity.scopes (admin bypass) (authorizer.ts)
4. session ownership: sessionId arg must be owned (ownership.ts)
5. session-creating tool → per-subject quota
allow → tool runs | deny → error result, tool skipped
──► afterCall(ctx, result)
bind newly-created sessions to caller; release deleted ones
Credential types (dispatched by shape in authenticator.ts):
ak_<id>_<secret>→ static API key (keystore, constant-time) — machines/CIst_<secret>→ session token minted byauth_login<h>.<p>.<s>JWT → OAuth access token verified via JWKS — humans/IDE
File map
| File | Responsibility |
|---|---|
src/plugin/auth-plugin.ts |
The plugin: register (auth_login/whoami/logout), beforeCall, afterCall, lifecycle |
src/plugin/create-plugin.ts |
Wire deps → createAuthPlugin / createAuthPluginFromEnv |
src/auth/authenticator.ts |
Dispatch a raw credential to the right verifier → Identity |
src/auth/keystore.ts |
API-key verify (hashed, timing-safe) + issued session tokens |
src/auth/oauth.ts |
OAuth JWT validation via jose (JWKS, iss/aud/exp) |
src/auth/authorizer.ts |
Scope/role checks per tool; admin role bypass |
src/auth/ownership.ts |
subject ⇄ Appium sessionId registry (multi-tenant isolation) |
src/auth/rate-limiter.ts |
Per-subject fixed-window limiter |
src/auth/keygen.ts |
API-key generation (keygen command): bearer + hash + config record |
src/core/audit.ts |
JSON-line audit events to stderr (never logs credentials) |
src/core/config.ts |
Env → AuthConfig (loadConfig), sha256Hex |
src/core/types.ts |
Identity, AuthConfig, ApiKeyRecord, Decision, etc. |
src/server/server.ts |
buildAuthenticatedServer = createAppiumMcpServer({ plugins:[auth] }) |
src/server/gateway.ts |
Credential-injecting reverse proxy for header-based clients (--gateway) |
src/server/http-credential.ts |
Pure helpers: read credential from headers, inject authToken into tools/call bodies |
src/server/credential-schema.ts |
Keep authToken from being stripped by tool schemas before beforeCall runs |
src/server/fastmcp-class.ts |
Resolve the exact FastMCP class instance that appium-mcp uses (for patching) |
src/index.ts |
Public package exports |
src/cli.ts |
Bin entry: keygen command; --gateway (header auth), --httpStream (arg auth), or stdio |
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 · 123 lines · 1,802 tokens per session scan A 3761d7437a01
appium-mcp-auth CLAUDE.md is an instructions file published in the GitHub repository appclawhq/appium-mcp-auth (5 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 1,802 tokens to every session, about $0.0090 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.
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.