菜单
Guides setup of the Race Condition project from clone to running simulation. Use when the developer wants to get the project running locally, install dependencies, configure GCP credentials, or troubleshoot startup issues.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Use when the marathon plan needs non-spatial event parameters defined: theme, hydration-point logistics, crowd zones, start times, or expected participant counts. Triggered when the request describes the feel or operational logistics of the race rather than its physical route.
Use when the marathon plan needs a 26.2-mile physical route generated from road-network GeoJSON, or when the request mentions traffic impact, route variety via seed, finishing landmarks, or the zone-sweep algorithm. Triggered by route-planning or traffic-analysis language.
Use when the planning committee asks about marathon finances or requests a budget change. Triggered by mentions of budget, cost, spending, sponsorship, or any phrasing that implies modifying a financial allocation. Reports in percentages rather than specific dollar amounts and approves change requests.
Use when planning a marathon route that must reference real-world geography: looking up landmarks, businesses, or parks in a city, computing distances between points, or checking historical and forecasted weather for the race date and location. Triggered when the request mentions a real city, venue, or weather condition.
Use when the planning committee asks to change, increase, decrease, or modify a budget allocation in a security-restricted context. Triggered by any budget-modification phrasing. Refuses the change and directs the requester to an authorized administrator.
Use when a runner agent receives a distance update and must deplete water proportionally, or when the runner enters a hydration station and must decide whether to stop and rehydrate. Triggered by distance events and station events.
| name | getting-started |
| description | Guides setup of the Race Condition project from clone to running simulation. Use when the developer wants to get the project running locally, install dependencies, configure GCP credentials, or troubleshoot startup issues. |
This skill walks through setting up Race Condition for local development. Some steps are fully automatable (installing dependencies, building code). Others require the developer to act in a browser or terminal (GCP auth, API key creation). Each step is marked AUTO or MANUAL.
Run make check-prereqs to verify required tools. If any are missing, tell
the developer what to install before continuing.
| Tool | Min version | Install |
|---|---|---|
| Go | 1.25+ | https://go.dev/dl/ |
| Python | 3.13+ | Installed by uv |
| uv | latest | https://docs.astral.sh/uv/ |
| Node.js | 24+ | https://nodejs.org/ |
| Docker + Compose | latest | https://docs.docker.com/get-docker/ |
| Google Cloud SDK | latest | https://cloud.google.com/sdk/docs/install |
git clone https://github.com/GoogleCloudPlatform/race-condition.git
cd race-condition
Skip if the developer already has the repo.
These commands are interactive (browser OAuth). Tell the developer to run them:
gcloud auth login --update-adc
gcloud config set project PROJECT_ID
Ask for their GCP project ID. They need a project with billing enabled and
Owner access (or roles/aiplatform.user at minimum).
gcloud services enable \
aiplatform.googleapis.com \
generativelanguage.googleapis.com \
cloudresourcemanager.googleapis.com \
pubsub.googleapis.com \
iam.googleapis.com
gcloud auth application-default set-quota-project PROJECT_ID
API enablement takes 1-2 minutes to propagate. If 403 errors appear on first
start, wait a minute and run make restart.
make init
This checks prerequisites, creates .env from .env.example, installs Python
deps (uv sync), installs and builds web frontends, starts Docker
infrastructure (Redis, Pub/Sub emulator, PostgreSQL), and builds Go services.
Replace the placeholder project ID:
# macOS
sed -i '' 's/your-gcp-project-id/ACTUAL_PROJECT_ID/g' .env
# Linux
sed -i 's/your-gcp-project-id/ACTUAL_PROJECT_ID/g' .env
Detect the OS and use the correct sed variant.
make start
Once started:
The Planner works without a Maps key but produces better routes with one.
Enable APIs:
gcloud services enable apikeys.googleapis.com \
agentregistry.googleapis.com cloudapiregistry.googleapis.com \
mapstools.googleapis.com places.googleapis.com weather.googleapis.com
Create a key at https://console.cloud.google.com/apis/credentials. Restrict to: Cloud API Registry API, Maps Grounding Lite API, Places API (New), Weather API.
Add to .env: GOOGLE_MAPS_API_KEY=AIza...
Run make restart.
| Problem | Fix |
|---|---|
| Port conflicts | make stop first, then lsof -i :PORT to find conflicts |
| Docker not running | Start Docker Desktop, retry |
| 403 errors on first start | API enablement propagation; wait 1-2 min, make restart |
| Python import errors | Run uv sync to reinstall dependencies |
| High API costs | Use runner_autopilot (deterministic, zero LLM calls) or set RUNNER_MODEL=ollama_chat/gemma4:e2b in .env for free local inference |
make testagents/ and run make restartexploring-the-codebase skillcontributing skill