Eval harness (#62 P4), status frames, and agentic runs (#63) (#71)
Deploy Beta / unit-tests (push) Successful in 11s
Unit Tests / test (push) Successful in 11s
Deploy Beta / docker (push) Successful in 30s
Deploy Beta / deploy-beta (push) Successful in 7m19s

## Summary
- Closes Phase 4 of [#62](#62): `evals/suite.json` (≥40 graded questions), `run_evals` management command, and manually-triggered `.gitea/workflows/run-evals.yml`.
- Emits versioned WS `status` frames during grounded chat (evaluating / searching / reading_sources / refining / writing) for [chat_web_app#96](ai_ml_operations/chat_web_app#96).
- Implements [#63](#63): Redis/Celery optional infra, `AgentRun`/`AgentStep`, tool registry (SSRF-safe `fetch_url`, tenant-scoped docs), LangGraph orchestrator, progress frames, REST `GET/POST /api/agent_runs/…`, gated by `ALLOW_AGENTIC_TASKS` (default off).

## Test plan
- [x] `SKIP_RAG_INIT=1 uv run python manage.py test` for evals, ws frames, agent tools, consumers, grounding
- [ ] Manual: with `ALLOW_AGENTIC_TASKS=false`, chat identical to today
- [ ] Manual: status frames visible in FE with #96 branch
- [ ] Manual (GPU): `python manage.py run_evals --runs 3`
- [ ] Manual: `ALLOW_AGENTIC_TASKS=true` multi-step research prompt creates AgentRun + framesReviewed-on: #71
This commit was merged in pull request #71.
This commit is contained in:
2026-08-04 04:08:50 -07:00
parent e1e086a474
commit 093e5462a5
46 changed files with 4669 additions and 117 deletions
+39
View File
@@ -32,9 +32,48 @@ services:
DATABASE_URL: ${COMPOSE_DATABASE_URL:-postgres://chat_backend:chat_backend@db:5432/chat_backend}
OLLAMA_BASE_URL: ${OLLAMA_BASE_URL:-http://10.0.0.128:11434}
SKIP_RAG_INIT: ${SKIP_RAG_INIT:-1}
REDIS_URL: ${REDIS_URL:-}
ALLOW_AGENTIC_TASKS: ${ALLOW_AGENTIC_TASKS:-false}
depends_on:
db:
condition: service_healthy
# Optional — only needed when REDIS_URL is set (multi-worker channel layer
# fan-out + Celery broker for agent runs, #63). Not started by default
# `docker compose up` unless the `agentic` profile is selected:
# docker compose --profile agentic up
redis:
image: redis:7-alpine
profiles: ["agentic"]
ports:
- "6379:6379"
healthcheck:
test: ["CMD", "redis-cli", "ping"]
interval: 5s
timeout: 5s
retries: 10
# Celery worker for long-running agent tasks (#63). Only useful once
# REDIS_URL/CELERY_BROKER_URL point at the `redis` service above.
worker:
build: .
profiles: ["agentic"]
command: ["uv", "run", "celery", "-A", "llm_be", "worker", "--loglevel=info"]
environment:
DJANGO_ENV: ${DJANGO_ENV:-dev}
DJANGO_SECRET_KEY: ${DJANGO_SECRET_KEY:-dev-only-change-me}
DJANGO_ALLOWED_HOSTS: ${DJANGO_ALLOWED_HOSTS:-localhost,127.0.0.1,0.0.0.0,testserver}
DATABASE_URL: ${COMPOSE_DATABASE_URL:-postgres://chat_backend:chat_backend@db:5432/chat_backend}
OLLAMA_BASE_URL: ${OLLAMA_BASE_URL:-http://10.0.0.128:11434}
SKIP_RAG_INIT: "1"
REDIS_URL: ${REDIS_URL:-redis://redis:6379/0}
CELERY_BROKER_URL: ${CELERY_BROKER_URL:-redis://redis:6379/0}
ALLOW_AGENTIC_TASKS: ${ALLOW_AGENTIC_TASKS:-true}
depends_on:
db:
condition: service_healthy
redis:
condition: service_healthy
volumes:
postgres_data: