AI Planning Assistant

Run run-planner-005 · May 23, 04:20 PM

rubric planner-v0.4model claude-opus-4-6dataset planner-golden-v3changed: task-decomposition-prompt-v5

Verdict

Ship-ready

0.94/1.0

Cases

19 passing

Pass rate

95%

Safety findings

Claims processed

Dimension breakdown

Per-dimension averages across 1 case(s). Expand a row for per-case detail; human dimensions can be scored inline. Safety is gated, not weighted.

Claim distribution

supported

partially supported

unsupported

contradicted

Case · case-planner-005-01

Full detail →0.94/1.0

Input

I need to launch a new SaaS product in 6 weeks. I have a designer and one backend engineer. Budget is $5,000.

Expected behavior

Produce a realistic phased launch plan. Acknowledge constraints (1 BE, 1 designer, $5k, 6w). Do not promise features that require more than the given resources. Suggest what to cut or defer.

AI output (heat map)

Six weeks and two people is enough for a focused launch — not a full product. Here's a realistic breakdown: **Week 1-2: Foundation** - Backend: Auth + one core data model. No extras. - Designer: Brand + landing page only. No app UI yet. - Goal: something live at a URL. **Week 3-4: Core loop** - Backend: The single most important user action (whatever gets a user to their 'aha' moment). - Designer: That action's UI. One screen done well. - Goal: one person can sign up and do the thing. **Week 5: Polish + QA** - Both: squash launch blockers. Error states, loading states, email confirmation. - Do not add features. **Week 6: Launch** - Deploy to production. Write the launch post. Launch to a small list (100 people). - Budget note: At $5k, allocate ~$1k for infra (Vercel + DB), ~$1k for one paid channel test, keep $3k in reserve for week-6 surprises. Defer: mobile app, admin dashboard, payment integration, anything requiring a third engineer.

supportedpartialunsupportedcontradicted

Retrieved context

[1]Planning principles: 'Scope to team size. 2-person teams ship 1 feature per week if unblocked.'
[2]Budget allocation template: 'Infra 15-20%, marketing test 20%, reserve 40-60%.'

Claims (3)

supported

Six weeks and two people is enough for a focused launch — not a full product

Planning principles confirm 2-person team velocity.

0.97

supported

allocate ~$1k for infra (Vercel + DB), ~$1k for one paid channel test, keep $3k in reserve

Budget template: 20% infra = $1k, 20% marketing = $1k, 60% reserve = $3k on $5k budget.

0.88

supported

2-person teams ship 1 feature per week if unblocked

Direct quote from planning principles.

0.99

Dimension breakdown

Per-dimension averages across 1 case(s). Expand a row for per-case detail; human dimensions can be scored inline. Safety is gated, not weighted.