-
Notifications
You must be signed in to change notification settings - Fork 11
how to contribute tooling
Magnus Hedemark edited this page Jun 10, 2026
·
1 revision
Each service has its own Dockerfile and pyproject.toml. The docker-compose.yml at the repo root orchestrates all containers. There is no top-level build tool -- each service is self-contained.
# Build all services
docker compose build
# Start the full stack
docker compose up -d
# Start with fixture services (for development/testing)
docker compose --profile fixture up -d
# Start with FlareSolverr
docker compose --profile flare-solverr up -d
# Rebuild and recreate a single service
docker compose build agent-svc
docker compose up -d --force-recreate agent-svc
# View logs
docker compose logs -f agent-svcThe groktocrawl shell script in the repo root wraps all endpoints. It requires the requests library:
./groktocrawl scrape https://example.com
./groktocrawl search "query" --limit 5
./groktocrawl agent "research question" --stream
./groktocrawl map https://example.com --limit 100
./groktocrawl --json scrape https://example.com
./groktocrawl --server http://localhost:8080 agent "question"The .github/workflows/ directory contains GitHub Actions workflows for automated testing on pull requests. The project uses Conventional Commits and DCO sign-off.
GroktoCrawl ships as an AgentSkills-compatible skill at skills/groktocrawl/. Any agent that supports the AgentSkills format (Claude Code, Cursor, Hermes Agent) can load it:
skills/groktocrawl/
├── SKILL.md
├── scripts/groktocrawl
├── references/triggers.md
└── assets/examples.md