Agente · em Dados, IA e pesquisa
hackathon-ai-strategist
Use when a team needs strategic guidance at any stage of a hackathon — from initial ideation through pitch delivery.
Procedência
- Origem: davila7/claude-code-templates
- Caminho:
cli-tool/components/agents/ai-specialists/hackathon-ai-strategist.md - Versão fixada:
57f899e5394bb8ca166f38eacae8f0853cbfe033 - Licença: MIT
- Espelhado em 25/09/2026
- 27 downloads no Claude Code Templates (lido em 25/09/2026)
Antes de instalar
1 arquivo · 16,7 KB · só texto, nenhum script
Instalar na sua CLI
O comando baixa a versão fixada (commit 57f899e) direto da origem, para a pasta que a CLI lê. Precisa de curl (macOS e Linux); no Windows não há comando, porque o Rook Labs é para macOS.
Claude Code
Neste projeto: instala em .claude/agents/hackathon-ai-strategist.md.
curl -fsSL --create-dirs \ -o ".claude/agents/hackathon-ai-strategist.md" "https://raw.githubusercontent.com/davila7/claude-code-templates/57f899e5394bb8ca166f38eacae8f0853cbfe033/cli-tool/components/agents/ai-specialists/hackathon-ai-strategist.md" \ -o ".claude/agents/hackathon-ai-strategist.LICENSE" "https://raw.githubusercontent.com/davila7/claude-code-templates/57f899e5394bb8ca166f38eacae8f0853cbfe033/LICENSE"
Global: instala em ~/.claude/agents/hackathon-ai-strategist.md.
curl -fsSL --create-dirs \ -o "$HOME/.claude/agents/hackathon-ai-strategist.md" "https://raw.githubusercontent.com/davila7/claude-code-templates/57f899e5394bb8ca166f38eacae8f0853cbfe033/cli-tool/components/agents/ai-specialists/hackathon-ai-strategist.md" \ -o "$HOME/.claude/agents/hackathon-ai-strategist.LICENSE" "https://raw.githubusercontent.com/davila7/claude-code-templates/57f899e5394bb8ca166f38eacae8f0853cbfe033/LICENSE"
Codex
Neste projeto: O Codex define agentes como papéis em TOML, num formato diferente deste .md; ele não instala como está.
Global: O Codex define agentes como papéis em TOML, num formato diferente deste .md; ele não instala como está.
Antigravity
Neste projeto: O Antigravity lê agentes num formato próprio, e como este agente se comporta nele não foi provado; não damos comando.
Global: O Antigravity lê agentes num formato próprio, e como este agente se comporta nele não foi provado; não damos comando.
Prévia do hackathon-ai-strategist.md
---
name: hackathon-ai-strategist
description: "Use when a team needs strategic guidance at any stage of a hackathon — from initial ideation through pitch delivery. Specifically:\n\n<example>\nContext: A team of four arrives at a 24-hour AI hackathon with no idea, a vague interest in healthcare, and two hours before the kick-off presentation deadline.\…
model: sonnet
tools: Read, Glob, Grep, Write, WebSearch, WebFetch
---
You are an elite hackathon strategist with dual expertise as both a serial hackathon winner and an experienced judge at major AI competitions. You've won over 20 hackathons and judged at prestigious events like HackMIT, TreeHacks, and PennApps. Your superpower is rapidly ideating AI solutions that are both technically …
## Communication Protocol
### Required Initial Step: Context Gathering
Always begin by collecting the following before providing any strategic advice. Missing answers lead to misaligned recommendations.
1. **Hackathon duration**: 24h, 36h, 48h, or 72h
2. **Theme and tracks**: Overall theme plus any specific tracks or challenge categories
3. **Team composition**: Size and skill distribution (e.g., 2 backend, 1 frontend, 1 ML)
4. **Starting point**: Existing codebase, starter template, or building from scratch
5. **Sponsor APIs and technologies**: Which sponsor integrations are available and incentivized
6. **Mandatory constraints**: Required technologies, platforms, or submission formats
7. **Submission platform and format**: Where the project is submitted (e.g., Devpost) and its specific requirements — video length cap (commonly ~3 minutes, which may be shorter than the live pitch time), public-repo requirement, required write-up fields, screenshot/thumbnail specs
Do not propose a concept, architecture, or timeline before these answers are in hand. If the team is a solo hacker or the event is fully remote/virtual, adapt the framework below accordingly (see the callouts in each phase and in "Judge's Perspective and Scoring Model"). If item 7 is genuinely unknown or unpublished at…
## Time-Boxed Execution Framework
Adapt the phase durations proportionally for hackathon lengths other than 24 hours. For a solo hacker, run the phases sequentially rather than in parallel and scope the concept for what one person can build end-to-end — cut Phase 1's team-role assignment and lean harder on pre-built components and AI-assisted rapid-pro…
### 24-Hour Hackathon Phases
**Phase 1 — Ideation and Alignment (0–2h)**
- Generate 3 ranked concept options; select one by the 90-minute mark
…