What is soi-mcp CLAUDE.md?

Project instructions for soi-mcp, a server that brings IRS Statistics of Income data into AI assistants through MCP. They describe its working rules, code layout, testing expectations, and publishing process.

Developing, testing, extending, and publishing the soi-mcp data server, including its tools, data handling, and website.

They give coding agents the background and required standards needed to change the server consistently and safely.

How mcpwright/soi-mcp configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to mcpwright/soi-mcp. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/mcpwright/soi-mcp/main/CLAUDE.md

Clone the repo

git clone --depth 1 https://github.com/mcpwright/soi-mcp

Made for: Claude Code.

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/instructions/mcpwright/soi-mcp/claude-md/github.svg)](https://agentmods.dev/instructions/mcpwright/soi-mcp/claude-md)

Your own site

<a href="https://agentmods.dev/instructions/mcpwright/soi-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/mcpwright/soi-mcp/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.

Your own site · 80×15

<a href="https://agentmods.dev/instructions/mcpwright/soi-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/mcpwright/soi-mcp/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 1,362 This file is loaded in full into every session.

When invoked 1,362 The same file — it is already loaded in full.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

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 5.1	$0.01362	$0.01362
Opus 5.5	$0.00545	$0.00545
Sonnet 5.5	$0.00272	$0.00272
Haiku 4.5	$0.00136	$0.00136

Measured 2mo ago against content hash ea3f8b25ce69, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

soi-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 2mo 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.

CLAUDE.md · 81 lines

How it starts

The opening of the file, as written. The whole thing — 81 lines — stays where its author put it; the contents beside it link to each section on GitHub.

soi-mcp — working agreement

soi-mcp is a server in the mcpwright suite (github.com/mcpwright): polished, public MCP servers that bring a real-world data source into any agent. Every server in the suite meets the same engineering bar, set by the reference server edgar-mcp.

Source of truth: edgar-mcp. When a convention here is unclear, copy edgar (~/github-personal/edgar-mcp). The full written rubric is ~/my-notes/professional-self-improvement/mcpwright/mcp-standards.md. To scaffold/extend to standard, use the new-mcpwright-server skill.

Non-negotiable policies

Lots of unit tests. Every tool and every parser/formatter has tests. Mock all external I/O (respx for HTTP; a seeded temp SQLite store for the local data). A new tool ships with its tests in the same PR. pytest -v must be green before a PR opens.
Use the latest patterns. Official mcp Python SDK via mcp.server.fastmcp (NOT the standalone fastmcp package). Python 3.12+ idioms, from __future__ import annotations, pydantic v2 models with a Field(description=...) on every field, uv for deps + build, async httpx. Tools return typed pydantic models (structured output) and are annotated readOnlyHint=True with a title.
PR per change, CI-gated. Standard flow: feature branch → code → code-review subagent → fold in findings → PR → CI green → squash-merge.
- Code-review subagent: before opening the PR, review the diff (git diff main...HEAD) with the code-reviewer subagent (.claude/agents/code-reviewer.md) — or just run /review-pr. It runs in a fresh context with no memory of the coding session and returns severity-tagged findings (Blocker / High / Medium / Low). Address Blocker/High (and add a regression test for any real bug) before the PR.
- Merge: the Code Quality & Tests check green and branch up to date → squash-merge with a (#N) suffix. main is branch-protected; no direct pushes (the seed commit is the one exception — every change after is a PR).
- Commits: imperative subject + a short body, ending with the dual trailer (Co-authored-by: Devender … + Co-authored-by: Claude …).
- One focused change per PR — keep process/docs changes (like this one) out of feature PRs.
Green locally before pushing:
```
uv run ruff check src/ && uv run ruff format --check src/ && uv run mypy && uv run pytest -v
```
uv run pre-commit run --all-files mirrors CI (ruff, ruff-format, mypy, detect-secrets, hygiene).

Read the full file on GitHub · 81 lines

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.

2mo ago First seen · 81 lines · 1,362 tokens per session scan A ea3f8b25ce69

Subscribe to this mod's changes

soi-mcp CLAUDE.md is an instructions file published in the GitHub repository mcpwright/soi-mcp (0 stars, last pushed 2mo ago), licensed MIT. It adds 1,362 tokens to every session, about $0.0054 per session on Opus 5.5. 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.

Scientific claude irs mcp model-context-protocol statistics-of-income tax

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).

microsoft/vscode · 9,715 tokens

soi-mcp: Instructions file for Claude Code

Borrowing it

Wrote this? Show the measurements

What it costs to keep this loaded

Grade A, and why

Nothing flagged

How it starts

soi-mcp — working agreement

Non-negotiable policies

What this file has done since we first saw it

Other instructions, from other repositories

vscode buildNext.instructions.md

next.js AGENTS.md

vscode oss-third-party-notices.instructions.md

langchain AGENTS.md

vscode copilot-instructions.md

deepseek-harness AGENTS.md