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/local-falcon/mcp/claude-mdgit clone --depth 1 https://github.com/local-falcon/mcpWrote 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/local-falcon/mcp/claude-md)<a href="https://agentmods.dev/instructions/local-falcon/mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/local-falcon/mcp/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 | $0.04856 | $0.04856 |
| Opus 5 | $0.02428 | $0.02428 |
| Sonnet 5 | $0.00971 | $0.00971 |
| Haiku 4.5 | $0.00486 | $0.00486 |
Grade A, and why
mcp 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 today.
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 — 383 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md — @local-falcon/mcp
Project Overview
This is the Local Falcon MCP Server (@local-falcon/mcp), a Model Context Protocol server that wraps the Local Falcon API. It enables AI agents to run geo-grid rank tracking scans, retrieve reports, manage campaigns, monitor Google Business Profiles, and analyze competitive positioning across Google Maps, Apple Maps, and AI search platforms.
Package: @local-falcon/mcp (npm)
Version: 1.4.9
License: MIT
Runtime: Node.js 18+
Language: TypeScript (strict mode)
Architecture
index.ts → Entry point. Transport selection (STDIO, SSE, HTTP), session management, OAuth 2.1
server.ts → MCP tool registrations. Exports getServer() which creates McpServer with 38 tools
localfalcon.ts → API client layer. All fetch functions, rate limiting, retry logic, timeout handling
oauth/ → OAuth 2.1 authorization server (routes, provider, config, state/client stores)
Key Design Patterns
server.tsexports a singlegetServer(sessionMapping)function that creates and returns anMcpServerinstance with all 38 tools registered viaserver.tool(name, description, zodSchema, annotations, handler). Every tool includes MCP tool annotations (readOnlyHint,destructiveHint) that signal to AI clients whether a tool reads data or modifies state.localfalcon.tscontains one exported function per API endpoint. Two call patterns:- URL params (v1):
new URL(endpoint)→url.searchParams.set()→ POST with JSON headers - FormData (v2):
new FormData()→form.append()→ POST with form body
- URL params (v1):
- API key resolution:
getApiKey(ctx)checks the session mapping first (for OAuth-authenticated remote sessions), then falls back toprocess.env.LOCAL_FALCON_API_KEY(for STDIO/local use). handleNullOrUndefined()converts null/undefined Zod outputs to empty strings before passing to API client functions. The API client functions then useif (value)guards to skip empty params.
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.
- today Changed 857133593fd8
- 5d ago First seen · 383 lines · 4,856 tokens per session scan A f0c2e61db814
mcp CLAUDE.md is an instructions file published in the GitHub repository local-falcon/mcp (23 stars, last pushed 2d ago), licensed MIT. It adds 4,856 tokens to every session, about $0.0243 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 2026-08-30.
Other instructions, from other repositories
gmgn-skills CLAUDE.md
Claude Code instructions for GMGNAI/gmgn-skills, covering claude.md, critical rule — read this first, project overview, available skills and quick decision guide.
the-seo-framework docs.instructions.md
Use when editing README.md, readme.txt, or other user-facing documentation for this repository. Covers readme wording constraints, WordPress readme format, changelog prose style, and public API naming requirements.
LocalAI CLAUDE.md
Claude Code instructions for mudler/LocalAI, a project described as: LocalAI is the open-source AI engine. Run any model - LLMs, vision, voice, image, video - on any hardware. No GPU required.
optifeed-radar CLAUDE.md
Instructions for optifeed/optifeed-radar, covering claude.md - optifeed radar, commands, stack, layout and hard rules (violations = rejected pr).
cutover-mcp-public AGENTS.md
AGENTS.md instructions for gocutover/cutover-mcp-public, covering agents.md, what this is, prerequisites, setup and environment variables.
cryptocurrency.cv CLAUDE.md
Claude Code instructions for nirholas/cryptocurrency.cv, covering cryptocurrency.cv (free crypto news), important rules, git identity & commits and terminal management.