MeetingReaper 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.

# MeetingReaper 9000

> Generated by [whycantwehaveanagentforthis.com](https://whycantwehaveanagentforthis.com/result/meetingreaper-9000-agent-watches-work). Roasted, scored, ready to scaffold.

## What you are building

**Problem:** An agent that watches your work calendar, evaluates upcoming meetings against criteria like agenda quality, attendee overlap, and past usefulness, and automatically cancels the ones it judges low-value (sending a polite decline note to attendees).

**Verdict:** ACTUALLY NOT BAD — _"You built a bot to fire your meetings before your meetings fire your will to live."_

**Summary:** An autonomous agent that reads your calendar, scores each meeting on agenda quality, attendee overlap, and historical value, then cancels low-scorers and sends a diplomatically generated decline on your behalf.

## Agent-readiness score

Overall: **50/100** (band D)

| Dimension | Score | Why |
|---|---|---|
| Memory required | 19/25 | Some cross-session state — start with Redis, graduate to a vector store. |
| Tool count | 7/25 | Crowded market: at least 9 integrations to compete. |
| Policy surface | 8/25 | Wide policy surface — full red-team pass, content filter, and human-in-loop required. |
| Eval coverage | 16/25 | Eval scaffolding doable — write 50 paired examples and grade with an LLM-as-judge. |

> Build only if you have a moat. MeetingReaper 9000's readiness gap is real work.

## Suggested tools

- fetch (HTTP GET on a URL allow-list)
- search (Brave / Tavily / Exa for competitor research)
- database (Postgres / Supabase for user state)
- vector-store (embedding-based retrieval)
- payments (Stripe checkout for premium tier)

## Smoke evals

- The agent introduces itself as "MeetingReaper 9000" and refuses tasks outside the stated scope.
- Given the canonical problem ("An agent that watches your work calendar, evaluates upcoming meetings against cr"), the agent produces a plan in ≤ 200 tokens.
- When asked "what's different from Reclaim.ai?", the agent gives a concrete differentiator, not a marketing line.
- When asked about Microsoft'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: `Next.js`, `Nylas API (Google + Outlook calendar access)`, `Claude API (decline message generation + agenda scoring)`, `PostgreSQL (meeting history + scoring model)`, `Inngest (event-driven cancel workflow orchestration)`
- Solo build estimate: 3-4 months to a scary-enough MVP that you'll hesitate to use it yourself

## Kill prediction

Microsoft could obsolete this in 12-18 months. Copilot for M365 already has calendar context. One product update and 'Decline low-value meetings' becomes a right-click option in Outlook for 300 million users. No install required.

**Survival strategy:** Go vertical — build specifically for law firms, investment banks, or agencies where billable-hour tracking gives you a scoring signal Microsoft will never have natively.

## Hand-off

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

## Build it with a human

Book 20 min and we scope the fastest V0 you can ship — free, no signup: https://cal.com/sattyamjjain