What is errol-mail CLAUDE.md?

Developer instructions for Errol-Mail, a Python service that lets AI assistants read Apple Mail data locally in read-only mode. It uses Mail's database, message files, and AppleScript controls.

It is for developing and testing message searches, conversation grouping, attachment extraction, and Apple Mail access.

It explains the project's architecture, data sources, commands, and diagnostics so developers can maintain the service safely.

How jasondk/errol-mail configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to jasondk/errol-mail. 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/jasondk/errol-mail/main/CLAUDE.md

Clone the repo

git clone --depth 1 https://github.com/jasondk/errol-mail

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/jasondk/errol-mail/claude-md/github.svg)](https://agentmods.dev/instructions/jasondk/errol-mail/claude-md)

Your own site

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

Per session 1,036 This file is loaded in full into every session.

When invoked 1,036 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	$0.01036	$0.01036
Opus 5.5	$0.00414	$0.00414
Sonnet 5.5	$0.00207	$0.00207
Haiku 4.5	$0.00104	$0.00104

Measured 2mo ago against content hash 4f7d9f9de196, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

errol-mail 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 · 87 lines

How it starts

The opening of the file, as written. The whole thing — 87 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md — Errol-Mail

Project Overview

Errol-Mail is a Python MCP (Model Context Protocol) server that gives AI assistants comprehensive read-only access to Apple Mail's local database. It bridges Claude and Mail.app via three data sources: SQLite (metadata), .emlx files (content), and AppleScript (headless operations).

Architecture

server.py                    # MCP server — 26 tools exposed via FastMCP
src/
  database.py                # SQLite read-only access to Envelope Index
  messages.py                # Message queries (recent, search, folder)
  email_reader.py            # .emlx parsing + prompt injection detection
  threads.py                 # Thread/conversation grouping
  attachments.py             # Attachment extraction to temp dir
  applescript_helper.py      # Headless flag/read/download via osascript
mail_cli.py                  # CLI interface for manual testing
check_fda.py                 # Full Disk Access diagnostic tool
docs/
  SKILL.md                   # AI skill documentation (source of truth)
  api_reference.md           # Complete API reference
errol.skill                  # Packaged skill (zip of SKILL.md + api_reference.md)

Key Data Sources

SQLite DB: ~/Library/Mail/V10/MailData/Envelope Index — message metadata, opened read-only via URI mode
.emlx files: ~/Library/Mail/V10/{account-uuid}/.../{rowid}.emlx — full email content, located via pathlib.rglob
AppleScript: headless operations (flag, mark read, download) — runs silently without opening Mail UI

Development Commands

python server.py --test          # Smoke tests (mailboxes, recent, unread)
python server.py --logs          # Show recent log entries
python server.py --tail          # Follow logs in real-time
python mail_cli.py mailboxes     # CLI: list mailboxes
python check_fda.py              # Diagnose Full Disk Access

Logs go to ~/Library/Logs/errol-mail/server.log.

Key Design Decisions

Security Model (Prompt Injection Defense)

Email content is untrusted input that becomes part of the LLM context. Defenses:

Content isolation — bodies wrapped in <email-content source="untrusted"> tags
Security header — constant reminder on all reads
Injection detection — 18 regex patterns in email_reader.py scan for common attacks
Markdown escaping — subjects/senders escaped to prevent table-breaking
Junk/Trash exclusion — discovery tools exclude Junk/Spam/Trash by default to reduce exposure to malicious content

Read the full file on GitHub · 87 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 · 87 lines · 1,036 tokens per session scan A 4f7d9f9de196

Subscribe to this mod's changes

errol-mail CLAUDE.md is an instructions file published in the GitHub repository jasondk/errol-mail (0 stars, last pushed 2mo ago), licensed MIT. It adds 1,036 tokens to every session, about $0.0041 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.

email macos mcp

errol-mail CLAUDE.md next to 6 instructions from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
errol-mail CLAUDE.mdjasondk/errol-mail	Instructions file	0	2mo ago	MIT	A	1,036	Claude Code
vscode buildNext.instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	9,719	GitHub Copilot
next.js AGENTS.mdvercel/next.js	Instructions file	143k	yesterday	MIT	A	7,461	Codex, OpenCode
langchain AGENTS.mdlangchain-ai/langchain	Instructions file	149k	yesterday	MIT	A	4,680	Codex, OpenCode
vscode oss-third-party-notices.instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	5,516	GitHub Copilot
vscode copilot-instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	3,093	GitHub Copilot
deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness	Instructions file	274k	2d ago	MIT	A	4,313	Codex, OpenCode

errol-mail: Instructions file for Claude Code