feat: bind all agent slots to active persona model routing
This commit is contained in:
parent
b9e4938a0f
commit
98f25c5848
23 changed files with 573 additions and 42 deletions
5
server-tools/persona-model-smart-router/README.md
Normal file
5
server-tools/persona-model-smart-router/README.md
Normal file
|
|
@ -0,0 +1,5 @@
|
|||
# OpenLux cost-first persona model router
|
||||
|
||||
The router uses the OpenAI-compatible endpoint at `https://api.openlux.ai/v1`. Its API key is read at call time from macOS Keychain service `ai.openlux.api`, account `guanghu-persona-agent-router`; no secret is stored in this repository.
|
||||
|
||||
Each call must carry the active parent persona, derived child Agent ID, controller, channel, job role, bounded authority and task. The router tries the configured low-cost healthy models first and escalates only after a transport, empty-response or downstream validation failure. Provider pricing is not exposed by the current `/models` response, so the configured order is explicitly a heuristic rather than a verified price table.
|
||||
Loading…
Reference in a new issue