AI-Generated

“An agent that catalogues my emails”

InboxArchivistBot 9000

ALREADY EXISTS, YOU'RE LATE
2/10
“Congratulations, you just reinvented Gmail labels, a feature that shipped in 2004.”

An AI agent that reads incoming emails, extracts metadata, and organizes them into categories, folders, or a searchable database automatically.

This is one of the most over-built categories in all of software. Every major email client has native cataloguing. The AI-native wave already produced SaneBox, Shortwave, and Superhuman. Building another one is like opening a new search engine in 2003 — theoretically possible, practically a cry for help.

whycantwehaveanagentforthis.com
Try Your Own Problem

Viability Analysis

Market Demand55
Tech Feasibility95
Competition92
Monetization30
AI Disruption Risk95
Fun Factor35

Pros & Cons

What's going for it

Easy to build an MVP in a weekend using Gmail API + Claude — great portfolio project
Clear, well-documented APIs from Google and Microsoft make integration straightforward
Personal use case is immediately satisfying — you'd actually use the thing you built
Could find a niche in enterprise compliance logging where generic tools fall short

What's against it

Gmail, Outlook, and Apple Mail already do this natively for free — your TAM is people who can't click 'Create Filter'
OAuth email permissions scare users into abandonment — conversion rates are brutal in this category
SaneBox has 15 years of training data on email patterns; you're starting from zero
Google's Gemini is being baked directly into Gmail — your standalone tool becomes redundant before launch
Email data is deeply sensitive — one breach and you're in every tech journalist's 'startup horror story' roundup

Who You're Up Against

SaneBoxsaas

AI-powered email filtering and folder organization that's been doing this since 2010

high
Superhumansaas

AI email client with auto-categorization, triage, and summarization for $30/month

high
Shortwavesaas

Google-backed AI email app with automatic bundling and cataloguing

high
Astro Mailsaas

AI email assistant acquired by Slack in 2019, then immediately killed — a cautionary tale

low
Mailstromsaas

Email organization and bulk management tool — still limping along

medium

Open Source Alternatives

imaplib + scikit-learn email classifieropen source

Dozens of Python repos on GitHub doing ML-based email classification — your idea is a homework assignment

medium
n8n Email Workflow Templatesopen source

No-code automation platform with pre-built email cataloguing workflows anyone can run for free

high
MailMateopen source

Power-user email client for Mac with deep rule-based cataloguing and smart mailboxes

medium

When Will Big AI Kill This?

Most Likely Killer

Google

Timeline: Already happening — Gemini in Gmail launched in 2024

Now3mo6mo1yr2yrNever

How They'll Do It

Google is embedding Gemini directly into Gmail with native summarization, categorization, and smart replies. No OAuth dance, no third-party trust issues, zero extra cost for Workspace users. Your agent doesn't even get a chance to load.

Your Survival Strategy

Niche down hard into a specific vertical — legal firms needing compliance-grade email logging, or researchers needing semantic search across years of academic correspondence. Generic email cataloguing is dead. Domain-specific cataloguing with specialized ontologies might survive.

Confidence

97%

If You're Crazy Enough to Build It

Solo Dev Time

1-2 weekends if you've touched an API before

Team Size

1 developer who really should just use SaneBox

Estimated Cost

$200-800 to build, $50-200/month to run at small scale

Tech Stack

Gmail API / Microsoft Graph APIClaude API or GPT-4oNext.jsPostgreSQLResend for notifications
How this was generated
29%PLAUSIBLE

Production-readiness odds

Worth pursuing — but expect the production gap to be the long pole, not the prototype.

ANCHORED TO OUR OWN READINESS RUBRIC — NO EXTERNAL STAT CITED

🛡 Safety considerations

What these mean →

Heuristic, not exhaustive. Surfaces the 3 biggest categories an operator should think about for this idea. Hover any chip for the mitigation pointer.

⚖ Governance checklist

5 controls apply

Things to have in place before you ship. Pairs with the OWASP-style risk chips above — that catalog answers “what could go wrong?”, this one answers “what should you have ready?”

  • Audit trail of every tool call

    critical

    Persist a structured per-call log of inputs, outputs, and decisions for at least the legal retention window. Without this, post-incident review is impossible.

  • Secrets management

    high

    Tokens and API keys live in a vault, not in env vars on a CI runner. Rotate on a documented schedule, not "when something happens."

  • Eval coverage on every release

    high

    A frozen eval suite that runs on every model / prompt change. "It worked when I demoed it" is not a release gate.

  • Per-user / per-tenant rate limits

    medium

    Agent loops are pathologically expensive when wrong. Cap tokens-per-session, tool-calls-per-session, and dollars-per-day before launch.

  • Pin model versions; track the changelog

    medium

    A silent provider-side model upgrade can shift behavior overnight. Pin to a versioned model ID; subscribe to the provider changelog.

OUR INTERNAL TWELVE-CONTROL SYNTHESIS — STANDARD SOC 2 / ISO 27001 / GDPR FAMILIES APPLIED TO LLM AGENTS

Agent-Readiness Score

Ready to scaffold today. InboxArchivistBot 9000 could be a working prototype in a week.

73BAND B
  • Memory ↗23/25

    Stateless or single-session — minimal memory layer.

  • Tools ↗11/25

    Crowded market: at least 8 integrations to compete.

  • Policy ↗15/25

    Mid-size policy surface — define refusal categories before launch.

  • Evals ↗24/25

    Established eval pattern — golden datasets and public benchmarks already exist.

↓ Download policy YAMLOpen in Managed Agents →Clone in Cursor

DETERMINISTIC SCORE — DERIVED FROM EXISTING ANALYSIS, NO SECOND LLM CALL

🛠 Build this with Claude Code

Skip the boilerplate. Start from a working spec.

We've packaged this idea into a CLAUDE.md + scaffold.sh starter — the problem statement, agent-readiness sub-scores, suggested tools, and smoke evals, all deterministic and ready to drop into a fresh repo. Open it in Claude Code, or copy the markdown into any IDE.

↓ Download CLAUDE.mdOpen in Claude Code →

Don't have Claude Code yet? View the bootstrap preview · grab the JSON bundle · or embed the readiness badge.

Want to actually build this?

Work with me to ship it.

Survived the verdict? Good. Let's build the damn thing.

Book a 30-min call
Report this content

Got another problem that needs an agent?

Roast My Problem

whycantwehaveanagentforthis.com