jbergner
a6bc71fb3a
Some checks failed
release-tag / Resolve release metadata (push) Successful in 30s
release-tag / Build knowledge (push) Failing after 4m51s
release-tag / Build control (push) Failing after 5m0s
release-tag / Build agent (push) Failing after 5m0s
release-tag / Build agent-data-init (push) Failing after 5m5s
release-tag / Build neuroforge-worker (push) Failing after 5m7s
release-tag / Build neuroforge (push) Failing after 5m9s
116 lines
4.8 KiB
Markdown
116 lines
4.8 KiB
Markdown
# Migration Manifest
|
||
|
||
## NeuroForge
|
||
|
||
Neu/erweitert:
|
||
|
||
- `internal/httpapi/integration.go` – App-Key-geschützte Knowledge-/Event-Integrationsendpunkte
|
||
- `internal/httpapi/integration_api_test.go` – Lifecycle, Auth und Namespace-Isolation
|
||
- `internal/store/store.go` – provenance-gefilterte semantische Suche
|
||
- `internal/store/batch.go` – Batch-Delete zur Vermeidung mehrfacher ANN-Rebuilds
|
||
- `cmd/server/main.go` – gemeinsamer Ollama-Endpunkt per Mega-Environment
|
||
- vorhandene NFVJ2/SQAR-Vector-Journal-Migration bleibt Bestandteil der Plattform
|
||
|
||
## GLPI AI Agent
|
||
|
||
Neu/erweitert:
|
||
|
||
- `internal/knowledge/neuroforge_backend.go` – SemanticBackend + NeuroForge HTTP-Client
|
||
- `internal/knowledge/store.go` – Hybrid-Retrieval mit `local`/`dual`/`neuroforge`
|
||
- `internal/knowledge/persistent_index.go` – inkrementelle Sync-/Externalisierungslogik
|
||
- `internal/config/config.go` – kontrollierbare Backend-/Failure-Parameter
|
||
- `cmd/agent/main.go` – Backend-Wiring
|
||
- Tests für Remote-Evidenz und fail-open/fail-closed
|
||
|
||
## GLPI AI Knowledgebase
|
||
|
||
Neu/erweitert:
|
||
|
||
- `internal/staging/staging.go` – quellenbewusste Staging-Proposals
|
||
- `cmd/server/app.go` – `POST /api/integrations/staging`
|
||
- `cmd/server/main.go` – Health und Staging-Ingress ohne Weitergabe von Editor-Credentials
|
||
- Tests für Staging-only-Governance und Auth-Grenzen
|
||
|
||
## Mega Platform
|
||
|
||
Neu:
|
||
|
||
- `docker-compose.yml`
|
||
- `go.work`
|
||
- `.env.example`
|
||
- `services/control/` – read-only Control Center
|
||
- `scripts/validate.sh`
|
||
- `scripts/status.sh`
|
||
- `scripts/generate-secrets.sh`
|
||
- `scripts/propose-draft.sh`
|
||
- gemeinsame `knowledge/`, `staging/`, `backups/`
|
||
- Architektur-, Kontroll-, Cutover-, Betriebs- und Validierungsdokumentation
|
||
|
||
## Version 1.1.0 – GLPI Relations & Obsidian Export
|
||
|
||
- GLPI-KB-Sync liest verfügbare `KnowbaseItem_Item`-Relationen über die installierte OpenAPI.
|
||
- `linked_items` werden im Agent-Knowledge-Modell erhalten.
|
||
- Knowledgebase und Agent exportieren Obsidian-kompatible ZIP-Vaults.
|
||
- Exporte enthalten YAML-Frontmatter, Wikilinks, Schema, Index, Manifest und `graph.json`.
|
||
- Neuer CLI-Helfer: `scripts/export-obsidian.sh`.
|
||
- Control-Center-Sicherheitsgrenze und delegierte Interaktionswege sind in `docs/CONTROL-CENTER.md` dokumentiert.
|
||
- Repräsentativer Export-Snapshot liegt unter `exports/knowledge-obsidian-snapshot.zip`.
|
||
|
||
## Version 1.2.0 – Controlled Autonomy
|
||
|
||
### NeuroForge
|
||
|
||
- `internal/httpapi/outcomes.go` – schmaler App-Key-Pfad für menschlich validierte Ticket-Outcomes
|
||
- `internal/brain/brain.go` – interne, nicht vom JSON-Client spoofbare Trusted-Provenance-Felder
|
||
- `cmd/server/main.go` – Controlled-Learning- und Research-Bootstrap per Environment
|
||
- `deploy/learning-policy.example.json` – konservative Source-Trust-/Learning-Policy
|
||
|
||
### GLPI AI Agent
|
||
|
||
- `internal/learning/outcomes.go` – persistentes Outcome-Audit, Sync-Status, Revisionen und NeuroForge-Sink
|
||
- `internal/agent/agent.go` – Stale-Run-Prüfung und Outcome-gated Learning
|
||
- `internal/web/server.go` / Dashboard – Bestätigen/Korrigieren und Audit-Sicht
|
||
- neue Outcome-Learning-Konfiguration mit explizitem fail-open/fail-closed
|
||
|
||
### Mega Platform
|
||
|
||
- `searxng` als optionaler Compose-Profilservice `research`
|
||
- `deploy/searxng/settings.yml`
|
||
- `scripts/research-up.sh`
|
||
- separate Research-, SearXNG- und Autonomy-Schalter
|
||
- Control Center zeigt diese Betriebsmodi read-only
|
||
- `docs/CONTROLLED-AUTONOMY.md`
|
||
- `docs/MIGRATION-v1.1.0-to-v1.2.0.md`
|
||
- `RELEASE-NOTES-v1.2.0.md`
|
||
- `patches/v1.1.0-to-v1.2.0.diff`
|
||
|
||
## Version 1.3.0 – Closed Learning Loop
|
||
|
||
### NeuroForge
|
||
|
||
- `internal/store/source_index.go` – rebuildbarer Provenance-Source-Index und atomare Memory-Supersession
|
||
- `internal/store/store.go` – source-begrenzter Exact-Fallback statt Full-Catalog-Scan
|
||
- `internal/brain/brain.go` – Multi-Source-Outcome-Suche
|
||
- `internal/httpapi/outcomes.go` – aktive Outcome-Suche + Remote-Supersession
|
||
- `internal/httpapi/metrics.go` – NFVJ2/SQAR Savings-/Block-Metriken
|
||
- Tests für aktive Revision, supersedete Revision und Source-Index-Rebuild nach Neustart
|
||
|
||
### GLPI AI Agent
|
||
|
||
- `internal/learning/outcomes.go` – OutcomeRetriever über den schmalen NeuroForge-App-Key-Pfad
|
||
- `internal/agent/agent.go` – validierte Erfahrung als sekundärer Reply-Kontext + Learning/Retrieval-KPIs
|
||
- `internal/model/model.go` – auditierbare `ValidatedOutcomeEvidence`
|
||
- `internal/ollama/client.go` – Prompt-Guard: Outcome darf nur KB stützen/widerlegen, nie selbst autorisieren
|
||
- `internal/web/server.go` – read-only `/api/quality/replay` und Statusmetriken
|
||
- Dashboard zeigt verwendete Erfahrungen, Similarity, Dauer und Fehler
|
||
|
||
### Mega Platform
|
||
|
||
- `scripts/quality-replay.py`
|
||
- `docs/QUALITY-REPLAY.md`
|
||
- `docs/QUALITY-REPLAY-example.json`
|
||
- `docs/MIGRATION-v1.2.0-to-v1.3.0.md`
|
||
- `RELEASE-NOTES-v1.3.0.md`
|
||
- Control Center zeigt Outcome Retrieval und Replay-Verfügbarkeit read-only
|
||
- Upgrade-Patch: `patches/v1.2.0-to-v1.3.0.diff`
|