aaronrene / knowtation public
.env.example
32 lines 1.5 KB
Raw
sha256:8915fe406161f95c1681f9469375e7bae5b28c884f00bedbdef65e4b0cd0738d docs(flow): commit FLOW-V0-SPEC.md hygiene for 7A-INT merge Human 16 hours ago
1 # Knowtation Hub Bridge — local `npm start` (see README "Run locally").
2 # The bridge loads the repo ROOT `.env` first (`../../.env` from this folder).
3 # Copy these lines there, or `export` them in your shell before `npm start`.
4
5 # Required (bridge exits without these):
6 CANISTER_URL=https://YOUR_CANISTER_ID.icp0.io
7 SESSION_SECRET=replace-with-a-long-random-string-min-32-chars
8
9 # Local URLs (adjust if you use different ports):
10 HUB_BASE_URL=http://localhost:3341
11 HUB_UI_ORIGIN=http://localhost:3333
12 HUB_UI_PATH=/hub
13
14 # Optional — GitHub "Connect" from hosted Hub when gateway proxies here:
15 # GITHUB_CLIENT_ID=
16 # GITHUB_CLIENT_SECRET=
17
18 # Optional — embeddings for Re-index / Search (default is Ollama on localhost):
19 # EMBEDDING_PROVIDER=openai|ollama|voyage|deepinfra
20 # OPENAI_API_KEY=
21 # EMBEDDING_MODEL=text-embedding-3-small # or nomic-embed-text, voyage-4-lite, BAAI/bge-large-en-v1.5, etc.
22 # VOYAGE_API_KEY= # when EMBEDDING_PROVIDER=voyage
23 # DEEPINFRA_API_KEY= # when EMBEDDING_PROVIDER=deepinfra (single key also drives chat in lib/llm-complete.mjs)
24 # Switching provider or dimension requires a vault re-index.
25
26 # Optional — hosted memory consolidation (Session 10 / Stream 1).
27 # CONSOLIDATION_LLM_API_KEY falls back to OPENAI_API_KEY when not set.
28 # CONSOLIDATION_LLM_MODEL defaults to gpt-4o-mini.
29 # CONSOLIDATION_COST_CAP_USD caps bridge-side daily LLM spend per user (optional).
30 # CONSOLIDATION_LLM_API_KEY=
31 # CONSOLIDATION_LLM_MODEL=gpt-4o-mini
32 # CONSOLIDATION_COST_CAP_USD=
File History 1 commit
sha256:8915fe406161f95c1681f9469375e7bae5b28c884f00bedbdef65e4b0cd0738d docs(flow): commit FLOW-V0-SPEC.md hygiene for 7A-INT merge Human 16 hours ago