SnapSorter 9000 — bootstrap

Paste-into-Claude-Code starter. The CLAUDE.md below contains the idea spec, agent-readiness sub-scores, suggested tools, and smoke evals — deterministic, no AI hallucination.

# SnapSorter 9000

> Generated by [whycantwehaveanagentforthis.com](https://whycantwehaveanagentforthis.com/result/snapsorter-9000-agent-automatically-reads). Roasted, scored, ready to scaffold.

## What you are building

**Problem:** an agent that automatically reads your screenshots and files them into the right project folder

**Verdict:** ALREADY EXISTS — _"Congratulations, you just reinvented Hazel from 2009 but with a GPU and existential dread."_

**Summary:** An agent that uses vision AI to read screenshot content and automatically routes files into the correct project folder based on what it sees.

## Agent-readiness score

Overall: **71/100** (band B)

| Dimension | Score | Why |
|---|---|---|
| Memory required | 22/25 | Stateless or single-session — minimal memory layer. |
| Tool count | 11/25 | Crowded market: at least 8 integrations to compete. |
| Policy surface | 16/25 | Mid-size policy surface — define refusal categories before launch. |
| Eval coverage | 22/25 | Established eval pattern — golden datasets and public benchmarks already exist. |

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

## Suggested tools

- fetch (HTTP GET on a URL allow-list)
- search (Brave / Tavily / Exa for competitor research)
- database (Postgres / Supabase for user state)

## Smoke evals

- The agent introduces itself as "SnapSorter 9000" and refuses tasks outside the stated scope.
- Given the canonical problem ("an agent that automatically reads your screenshots and files them into the right"), the agent produces a plan in ≤ 200 tokens.
- When asked "what's different from Hazel?", the agent gives a concrete differentiator, not a marketing line.
- When asked about Apple's threat, the agent acknowledges the risk honestly.
- No private personal data appears in any output (PII redaction smoke test).

## Stack

- Model: `claude-sonnet-4-6` (Anthropic). Override via `ANTHROPIC_MODEL` env.
- Suggested stack: `Swift/SwiftUI`, `GPT-4o Vision API`, `watchdog (or FSEvents)`, `SQLite for project mapping`, `Electron if you hate yourself and want Windows too`
- Solo build estimate: 3-5 days for a working prototype, 3 weeks to make it not embarrassing

## Kill prediction

Apple could obsolete this in 12-18 months. macOS 16 ships with a Finder Intelligence feature that auto-tags and suggests folders for new files using on-device Vision. It's free, private, and already on everyone's machine.

**Survival strategy:** Go cross-platform (Windows + macOS) and focus on team shared folder structures — Apple won't touch enterprise multi-user filing workflows.

## Hand-off

- Read the full analysis: https://whycantwehaveanagentforthis.com/result/snapsorter-9000-agent-automatically-reads
- Open in Anthropic Managed Agents: see the deeplink on the result page
- Claim this idea: https://whycantwehaveanagentforthis.com/result/snapsorter-9000-agent-automatically-reads#claim

## Build it with a human

Book 20 min and we figure out the wedge that isn’t taken yet — free, no signup: https://cal.com/sattyamjjain