llm-gateway

rene/llm-gateway

Fork 0

Commit Graph

Author	SHA1	Message	Date
Rene Fichtmueller	0191c60b64	chore: commit deployed gateway state (dashboard, streaming, routing, bridges, cost-tracking) Live production state on Erik that had drifted from Gitea — deployed across several sessions but never committed. Excludes deploy/ecosystem.config.cjs (holds live tokens). - dashboard: passive usage-report endpoint, per-device entries, CEST timezone, cost-panel rounding - completion: SSE + HTTP/2 streaming - pipeline: routing-rules, request-scorer, external-providers (subscription bridges) - cost-tracking: tokenvault migration, cost-calculator, request-logger - infra: docker-compose bridge env, server/health/tls, deps	2026-06-05 20:23:33 +00:00
Rene Fichtmueller	3a00ff4d33	feat: initial llm-gateway implementation - Complete Fastify gateway with 8-stage pipeline - Circuit breaker (opossum) per model tier - Rate limiting per caller - Ban list validation (EN/DE/auto-detected) - TIP validator (SFF-8024, part numbers, wavelengths) - Prometheus metrics - pg-boss async queue - PostgreSQL audit log + review queue - 9 prompt templates (TIP, LinkedIn, ShieldX) - Learning engine scaffolding - Auto-learning: ban-list, few-shot, routing, prompt optimizer	2026-04-02 22:48:55 +02:00

Author

SHA1

Message

Date

Rene Fichtmueller

0191c60b64

chore: commit deployed gateway state (dashboard, streaming, routing, bridges, cost-tracking)

Live production state on Erik that had drifted from Gitea — deployed across several
sessions but never committed. Excludes deploy/ecosystem.config.cjs (holds live tokens).

- dashboard: passive usage-report endpoint, per-device entries, CEST timezone, cost-panel rounding
- completion: SSE + HTTP/2 streaming
- pipeline: routing-rules, request-scorer, external-providers (subscription bridges)
- cost-tracking: tokenvault migration, cost-calculator, request-logger
- infra: docker-compose bridge env, server/health/tls, deps

2026-06-05 20:23:33 +00:00

Rene Fichtmueller

3a00ff4d33

feat: initial llm-gateway implementation

- Complete Fastify gateway with 8-stage pipeline
- Circuit breaker (opossum) per model tier
- Rate limiting per caller
- Ban list validation (EN/DE/auto-detected)
- TIP validator (SFF-8024, part numbers, wavelengths)
- Prometheus metrics
- pg-boss async queue
- PostgreSQL audit log + review queue
- 9 prompt templates (TIP, LinkedIn, ShieldX)
- Learning engine scaffolding
- Auto-learning: ban-list, few-shot, routing, prompt optimizer

2026-04-02 22:48:55 +02:00

2 Commits