Files
Timmy-time-dashboard/docker-compose.yml
Alexander Whitestone 88fa2f78ef
Some checks failed
Tests / lint (pull_request) Failing after 29s
Tests / test (pull_request) Has been skipped
feat: integrate SearXNG + Crawl4AI as self-hosted search backend (#1282)
Add web_search() and scrape_url() tools backed by self-hosted SearXNG
and Crawl4AI services — no paid API key required.

Changes:
- src/config.py: add timmy_search_backend, search_url, crawl_url settings
- src/timmy/tools/search.py: new module with web_search() and scrape_url()
  tools; both degrade gracefully when backend is unavailable or disabled
- src/timmy/tools/_registry.py: register search tools in full toolkit and
  tool catalog (available in orchestrator + echo)
- src/timmy/tools/file_tools.py: add web_search + scrape_url to echo (research) toolkit
- src/timmy/tools/__init__.py: export web_search, scrape_url
- docker-compose.yml: add searxng and crawl4ai services (profile: search)
- docker/searxng/settings.yml: SearXNG base config
- tests/timmy/test_tools_search.py: 21 unit tests with mocked HTTP
- AGENTS.md: document Search Capability section

Fixes #1282

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-03-23 21:51:43 -04:00

166 lines
5.9 KiB
YAML

# ── Development Compose ───────────────────────────────────────────────────────
#
# Services
# dashboard FastAPI app (always on)
# celery-worker (behind 'celery' profile)
# openfang (behind 'openfang' profile)
#
# Usage
# make docker-build build the image
# make docker-up start dashboard
# make docker-down stop everything
# make docker-logs tail logs
#
# ── Security note ─────────────────────────────────────────────────────────
# Override user per-environment — see docker-compose.dev.yml / docker-compose.prod.yml
#
# ── Ollama host access ──────────────────────────────────────────────────────
# By default OLLAMA_URL points to http://host.docker.internal:11434 which
# reaches Ollama running on the Docker host (macOS/Windows native).
#
# Linux: The extra_hosts entry maps host.docker.internal → host-gateway,
# which resolves to the host IP on Docker 20.10+.
services:
# ── Dashboard (FastAPI) ──────────────────────────────────────────────────
dashboard:
build: .
image: timmy-time:latest
container_name: timmy-dashboard
user: "" # see security note above
ports:
- "8000:8000"
volumes:
- timmy-data:/app/data
- ./src:/app/src # live-reload: source changes reflect immediately
- ./static:/app/static # live-reload: CSS/asset changes reflect immediately
environment:
DEBUG: "true"
OLLAMA_URL: "${OLLAMA_URL:-http://host.docker.internal:11434}"
# Grok (xAI) — opt-in premium cloud backend
GROK_ENABLED: "${GROK_ENABLED:-false}"
XAI_API_KEY: "${XAI_API_KEY:-}"
GROK_DEFAULT_MODEL: "${GROK_DEFAULT_MODEL:-grok-3-fast}"
# Search backend (SearXNG + Crawl4AI) — set TIMMY_SEARCH_BACKEND=none to disable
TIMMY_SEARCH_BACKEND: "${TIMMY_SEARCH_BACKEND:-searxng}"
TIMMY_SEARCH_URL: "${TIMMY_SEARCH_URL:-http://searxng:8080}"
TIMMY_CRAWL_URL: "${TIMMY_CRAWL_URL:-http://crawl4ai:11235}"
extra_hosts:
- "host.docker.internal:host-gateway" # Linux: maps to host IP
networks:
- timmy-net
restart: unless-stopped
healthcheck:
test: ["CMD", "curl", "-f", "http://localhost:8000/health"]
interval: 30s
timeout: 5s
retries: 3
start_period: 30s
# ── Celery Worker — background task processing ──────────────────────────
celery-worker:
build: .
image: timmy-time:latest
container_name: timmy-celery-worker
user: ""
command: ["celery", "-A", "infrastructure.celery.app", "worker", "--loglevel=info", "--concurrency=2"]
volumes:
- timmy-data:/app/data
- ./src:/app/src
environment:
OLLAMA_URL: "${OLLAMA_URL:-http://host.docker.internal:11434}"
extra_hosts:
- "host.docker.internal:host-gateway"
networks:
- timmy-net
restart: unless-stopped
profiles:
- celery
# ── SearXNG — self-hosted meta-search engine ─────────────────────────
searxng:
image: searxng/searxng:latest
container_name: timmy-searxng
profiles:
- search
ports:
- "${SEARXNG_PORT:-8888}:8080"
environment:
SEARXNG_BASE_URL: "${SEARXNG_BASE_URL:-http://localhost:8888}"
volumes:
- ./docker/searxng:/etc/searxng:rw
networks:
- timmy-net
restart: unless-stopped
healthcheck:
test: ["CMD", "wget", "-qO-", "http://localhost:8080/healthz"]
interval: 30s
timeout: 5s
retries: 3
start_period: 20s
# ── Crawl4AI — self-hosted web scraper ────────────────────────────────
crawl4ai:
image: unclecode/crawl4ai:latest
container_name: timmy-crawl4ai
profiles:
- search
ports:
- "${CRAWL4AI_PORT:-11235}:11235"
environment:
CRAWL4AI_API_TOKEN: "${CRAWL4AI_API_TOKEN:-}"
volumes:
- timmy-data:/app/data
networks:
- timmy-net
restart: unless-stopped
healthcheck:
test: ["CMD", "curl", "-f", "http://localhost:11235/health"]
interval: 30s
timeout: 10s
retries: 3
start_period: 30s
# ── OpenFang — vendored agent runtime sidecar ────────────────────────────
openfang:
build:
context: .
dockerfile: docker/Dockerfile.openfang
image: timmy-openfang:latest
container_name: timmy-openfang
profiles:
- openfang
environment:
OLLAMA_URL: "${OLLAMA_URL:-http://host.docker.internal:11434}"
OPENFANG_DATA_DIR: "/app/data"
extra_hosts:
- "host.docker.internal:host-gateway"
volumes:
- openfang-data:/app/data
networks:
- timmy-net
restart: unless-stopped
healthcheck:
test: ["CMD", "curl", "-f", "http://localhost:8080/health"]
interval: 30s
timeout: 5s
retries: 3
start_period: 15s
# ── Volumes ──────────────────────────────────────────────────────────────────
volumes:
timmy-data:
driver: local
driver_opts:
type: none
o: bind
device: "${PWD}/data"
openfang-data:
driver: local
# ── Internal network ────────────────────────────────────────────────────────
networks:
timmy-net:
driver: bridge