Pitch Deck Reviewer

What it takes in

The inputs you provide on each run, validated before anything executes.

2 tasks, run in order.

Each task passes its result to the next. Here's what each does and what it produces.

1. 01
   Critique the deck slide by slide. Go through every slide in {{deckContent}} in order. For each slide give: what works (the strongest element, stated specifically), what is weak or missing (the gap a skeptical partner would notice), and a concrete rewrite suggestion (a specific line, number, or restructure: not generic advice). Calibrate expectations to {{fundingStage}} if provided, and to what {{targetInvestors}} would care about if provided. Then assess the narrative arc across the whole deck: problem, why now, why you, traction, ask: and name exactly where the arc is strong and where it breaks. Ground every point in what the slides actually say; do not invent traction, metrics, or market facts. Present readably, never JSON.

   Expected output

   A readable critique using headings and bullets, with no JSON and no code blocks. Slide-by-Slide Critique: one heading per slide (Slide 1, Slide 2, and so on, with the slide's topic), each with What Works, What's Weak or Missing, and Rewrite Suggestion (a concrete, specific fix). Then Narrative Arc Assessment: how the deck performs across problem, why now, why you, traction, and ask, naming exactly where the arc holds and where it breaks.
2. 02
   Simulate the partner meeting. Using the critique from Task 1 and the deck in {{deckContent}}, write the 8 hardest questions this deck will get at {{fundingStage}} (or at the stage the deck implies if {{fundingStage}} is not provided): the skeptical partner's questions, not softballs. For each question, write a strong suggested answer grounded in the deck's actual content; where the deck lacks the material for a strong answer, say so and state what the founder needs to add or prepare. Then give the top 3 changes that would most improve the deck's odds, ranked by impact, each with a one-line rationale. Do not fabricate traction or facts the deck does not contain. Present readably, never JSON.

   Expected output

   A readable section with no JSON and no code blocks. The 8 Hardest Questions: numbered 1 to 8, each with the question and a Suggested Answer grounded in the deck's content (or a note on what the founder must add or prepare where the deck cannot support a strong answer). Then Top 3 Changes: ranked 1 to 3 by impact, each with the specific change and a one-line rationale for why it most improves the deck's odds.