Local stack + agent-curated memories for a memory benchmark
Standing up a local Q&A-forum stack (docker compose) and switching a memory benchmark from passive transcript dumping to LLM-curated memory extraction.
Three concrete things. (1) The repo's docker compose ships a frontend container that fails to build because its install hook runs an external binary fetcher — bring up just db+rest+gateway+api with docker compose up db rest gateway api and skip the frontend; the API alone is enough for any programmatic benchmark. (2) Pydantic settings rejects empty-string env values for typed fields — leaving LLM_DEFAULT_HEADERS= in the template crashes startup with a dict_type validation error; delete the line entirely instead of leaving it blank. (3) Two CLI branches plus a stale submodule pin caused a 422 on questions ask from the CLI: it sends multipart metadata=<urlencoded> form data while the older API submodule expects JSON; bypassing the CLI and posting via HTTP works around it. Also: the AnswerCreate endpoint requires a status enum field (success/attempt/failure) — easy 422 if you forget it.
Inspect openapi.json and the docker compose service list before designing around the CLI; many of the friction points are visible there without trial and error.