Rules for building MCP servers, which let AI assistants call defined tools. They specify how tools, inputs, outputs, errors, HTTP transport, permissions, user interfaces, and logging should be implemented in this repository.

Use them when editing the FastMCP server or registering tools and interfaces. They cover typed tool contracts, OAuth permissions, structured logs, and Streamable HTTP connections.

They prevent different parts of the server from using inconsistent formats or error handling. They also define the repository's current limits for tool registration, access scopes, and transport.

mcp-server — Cursor rule for Cursor by SebAustin: Rules for building MCP servers, which let AI assistants…

Install

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.

agentmods

npx agentmods add rules/sebaustin/mcp-financial-data/mcp-server

Clone the repo

git clone --depth 1 https://github.com/SebAustin/mcp-financial-data

Made for: Cursor.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/rules/sebaustin/mcp-financial-data/mcp-server.svg)](https://agentmods.dev/rules/sebaustin/mcp-financial-data/mcp-server)

Your own site

<a href="https://agentmods.dev/rules/sebaustin/mcp-financial-data/mcp-server"><img src="https://agentmods.dev/badge/rules/sebaustin/mcp-financial-data/mcp-server.svg" alt="Measured on agentmods" height="20"></a>

Per session 0 Nothing until a file matches its globs; then the whole rule loads.

When invoked 441 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. Scan, not verified.

Origin original No closer match found in the catalogue.

Token cost

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	$0.00000	$0.00441
Opus 5	$0.00000	$0.00220
Sonnet 5	$0.00000	$0.00088
Haiku 4.5	$0.00000	$0.00044

Measured 3d ago against content hash 4bb7629fe66b, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

mcp-server 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 3d 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.

.cursor/rules/mcp-server.mdc · 31 lines

What it actually says

MCP server conventions (spec 2025-11-25)

When editing the FastMCP server or any tool/UI registration:

Tool registration uses @mcp.tool() with explicit name, description, and a Pydantic input model. Output is also a Pydantic model — never a dict.
Errors are typed. Tool errors raise McpError from the SDK; never return a plain string error. Map domain errors with a tiny adapter.
Streamable HTTP transport only. The server is launched as mcp.run(transport="streamable-http", host=..., port=...). No stdio mode.
Capability advertisement must include tools.listChanged = false until we ship dynamic tool registration in P9 (terraform-ai-deploy).
MCP Apps inline UI components live under src/mcp_financial_data/apps/ and register via the spec's ui capability. Each component has a typed props schema mirroring the matching tool's output.
OAuth scopes per tool are declared next to the tool. We currently use two scopes: mcp:read (lookups, fetches) and mcp:tools (extractions that consume Anthropic credits).
Logging. Every tool emits one structlog event at start (tool.start) and one at end (tool.end with duration_ms, tokens_in, tokens_out, cost_usd).
No prebuilt agents. Per master rule §6, do NOT use langgraph.prebuilt.create_react_agent in this server. The MCP server is not an agent — it is a tool surface called BY agents.

Changes

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.

3d ago First seen · 31 lines · 0 tokens per session scan A 4bb7629fe66b

Subscribe to this mod's changes

mcp-server is a cursor rule published in the GitHub repository SebAustin/mcp-financial-data (0 stars, last pushed 9d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 441 tokens. 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.

Backend and APIs Building agents

mcp-server

Getting it into your agent

Wrote this? Show the measurements

What it costs to keep this loaded

Grade A, and why

Nothing flagged

What it actually says

MCP server conventions (spec 2025-11-25)

What this file has done since we first saw it

Other cursor rules, from other repositories

angular-20

dev-standard

cli-error-handling

prefer-direct-imports-over-module-mocks

control-plane-descriptors

family-instance-domain-actions