coulomb/llm-connect
5
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Resolve the OpenRouter key from OpenBao before the environment
All checks were successful
CI Smoke / host-smoke (push) Successful in 0s
Details
CI Smoke / container-smoke (push) Successful in 7s
Details

Browse source 
auto tries platform/suppliers/openrouter/llm-connect, then the live
activity-core workload lane, then OPENROUTER_API_KEY. A denied token
falls back to the environment. LLM_CONNECT_KEY_SOURCE selects env or
openbao alone.

Assistant: grok
Assistant-Session: 01a11ab6-1534-7d63-8606-662b92a80ed6
This commit is contained in:
tegwick 2026-10-08 19:35:00 +02:00
parent 965c4b3386
commit e1ca67e509
4 changed files with 173 additions and 11 deletions
Download patch file Download diff file Expand all files Collapse all files

4
README.md
View file

@ -74,7 +74,9 @@ llm-connect balance --json
# Which model and quote: docs/image-model-selection.md
# Render the one hall entry that is about to be published. Do not backfill the corpus.
llm-connect image render-hall --hall ../hall-of-helix --entry entries/<file>.md
# Receipt JSON on stdout; image bytes in --output. Credentials stay in the env.
# Receipt JSON on stdout; image bytes in --output.
# The OpenRouter key is read from OpenBao when the caller token can, then
# from OPENROUTER_API_KEY. LLM_CONNECT_KEY_SOURCE=openbao uses OpenBao only.
llm-connect image render \
--prompt "precise technical illustration of a gear, no text" \
--model "<image-model>" \