Add eval harness (#62 P4), status frames (#96), and agentic runs (#63).
CI / test (pull_request) Successful in 11s
Unit Tests / test (pull_request) Successful in 10s

Ship the Phase 4 accuracy eval suite with a manual Gitea workflow, emit
versioned WS status frames during grounded chat, and introduce opt-in
agent infrastructure (Redis/Celery, AgentRun/Step, tools, orchestrator)
gated by ALLOW_AGENTIC_TASKS so default chat behaviour stays unchanged.
This commit is contained in:
2026-08-04 06:07:26 -05:00
parent e1e086a474
commit 9c0b648db3
46 changed files with 4669 additions and 117 deletions
+5
View File
@@ -49,6 +49,11 @@ dependencies = [
"python-dateutil==2.9.0.post0",
"pytz==2025.2",
"stripe>=12.0.0,<14.0.0",
# Agentic task execution (#63): Redis-backed channel layer for multi-worker
# progress fan-out, plus Celery for durable long-running agent runs.
"channels-redis==4.3.0",
"redis==6.4.0",
"celery==5.5.3",
]
[dependency-groups]