Commit Graph

47 Commits

Author SHA1 Message Date
rcourtman 303a8e2fe8 Remove Assistant handoff decision guidance 2026-05-15 16:57:25 +01:00
rcourtman 14d3284233 Remove Assistant prompt routing heuristics 2026-05-15 16:10:58 +01:00
rcourtman f2d9d2aba8 Split overgreedy "tools not supported" classifier into three causes
The Patrol runtime classifier collapsed three distinct upstream
conditions into one misleading "Selected model does not support
Patrol tools" message:

  1. Provider rejected the *value* Pulse sent for tool selection
     (e.g. DeepSeek's "deepseek-reasoner does not support this
     tool_choice" — the model accepts tools, just not the forced
     coercion). The DeepSeek fix in 46145df9 dodges the symptom by
     coercing to auto, but the original misclassification pointed
     operators at the wrong remediation for 33 days.
  2. Provider has no tool-capable endpoint available for the
     selected model (OpenRouter's "No endpoints found …" surfaces
     this when account-level provider/data filters exclude every
     tool-capable route).
  3. Model truly lacks tool calling (the literal "tools are not
     supported" / "tool calling" cases).

Each now has its own PatrolFailureCause, title, summary,
description, and recommendation. summarizePatrolRuntimeFailureDetail
mirrors the split. Helper predicates patrolToolChoiceValueRejected
and patrolNoToolCapableEndpoint encapsulate the substring matching.

The OpenRouter "No endpoints found" test fixture now correctly
classifies as no_tool_capable_endpoint instead of
model_unsupported_tools — fixture updates in
patrol_runtime_failure_test.go, patrol_assistant_handoff_test.go,
and ai_handler_test.go reflect the more accurate diagnostic.
New tests cover the tool_choice_rejected and generic
model_unsupported_tools paths explicitly.

The ai-runtime contract is updated to note the classifier-split
obligation alongside the existing transport-shape obligation.
2026-05-10 14:10:18 +01:00
rcourtman b5c8e00859 Preserve previous successful fix across regressions
Capture the prior InvestigationRecord.ProposedFix.Description into a
new Finding.PreviousResolvedFixSummary field at regression time, before
the InvestigationRecord is cleared. Without this capture the next
investigation starts from blank context whenever a finding regresses,
and operational memory of "what worked last time" is lost.

The summary propagates through:
- FindingsStore.Add regression branch (capture before clear)
- Finding.MarshalJSON / UnmarshalJSON (wire shape)
- Both Finding to UnifiedFinding conversion sites in router.go
- UnifiedFinding (struct + JSON shadow + Marshal/Unmarshal)
- UnifiedStore.AddFromAI update branch (non-empty overwrite)
- Assistant chat context as a "Previous Resolved Fix" line so the LLM
  sees what worked previously rather than blank-slate diagnosing each
  regression.

Adds a unit test that walks the full lifecycle (detect, resolve via
UpdateInvestigationRecord + ResolveWithReason, re-detect) and asserts
PreviousResolvedFixSummary is preserved while InvestigationRecord is
cleared, plus two chat-context tests covering the surfaces-when-set and
omits-when-empty cases. Adds a contract test pinning the canonical
"previous_resolved_fix_summary" JSON key. Updates the api-contracts,
ai-runtime, and the dependent agent-lifecycle, performance-and-
scalability, and storage-recovery contracts to pin the operational
memory propagation rule and its scope boundary.
2026-05-08 19:45:44 +01:00
rcourtman dd15d23a35 Hydrate Patrol requester in Assistant handoffs
- hydrate approval requester identity into backend handoff actions
- include requester provenance in refreshed finding action context
- document the backend requester authority boundary
2026-05-08 00:26:04 +01:00
rcourtman bbaa6c0a6c Cover Patrol handoff approval mode
- force non-empty Patrol finding handoffs into approval-required mode

- add direct helper coverage for finding, scoped, resource, action, and metadata handoffs

- align subsystem contracts with the backend-owned clamp
2026-05-07 23:08:56 +01:00
rcourtman d9331d39e4 Move Patrol run Assistant handoffs server-side
- rehydrate Patrol run context from Patrol history for chat requests and follow-up sessions
- send only browser-safe run metadata from the frontend
- classify and redact provider runtime failures in handoff prompts and briefings
2026-05-07 21:58:06 +01:00
rcourtman ff9cb5c133 Restore Patrol run Assistant session identity
Refs #1463
2026-05-07 15:53:59 +01:00
rcourtman 6b4e4d7fe6 Merge Patrol handoff context into Assistant requests 2026-05-07 11:58:02 +01:00
rcourtman a66b80ceb9 Carry structured Patrol actions into Assistant handoffs 2026-05-07 10:48:15 +01:00
rcourtman 7aaf76d03c Carry live approval lifecycle into Assistant handoffs 2026-05-07 10:25:14 +01:00
rcourtman 812c86692d Route Assistant handoffs through model context 2026-05-07 03:00:58 +01:00
rcourtman 2c913fac8e Clamp Patrol finding chat handoffs to approval mode 2026-05-07 01:31:45 +01:00
rcourtman bf0e4e16ac Align Assistant briefings with handoff actions 2026-05-07 00:38:42 +01:00
rcourtman b8bad16d80 Recover live Patrol approvals for Assistant handoffs 2026-05-07 00:31:44 +01:00
rcourtman 1fc66cf0cb Add attention reasons to Assistant briefings 2026-05-06 23:39:33 +01:00
rcourtman 6b51a15dfe Frame Assistant operator decisions 2026-05-06 23:30:11 +01:00
rcourtman eb88119a29 Add evidence cues to Assistant briefings 2026-05-06 23:17:33 +01:00
rcourtman c50f051e13 Add evidence cues to Assistant briefings 2026-05-06 23:10:32 +01:00
rcourtman 2fd85f62d3 Add related finding recency to Assistant handoffs 2026-05-06 22:50:18 +01:00
rcourtman 1f424523da Surface Assistant finding lifecycle context 2026-05-06 21:21:35 +01:00
rcourtman 6269bc2afa Enrich Assistant related finding handoffs 2026-05-06 21:14:09 +01:00
rcourtman 67974f2ada Add Assistant operator briefing handoff 2026-05-06 20:48:25 +01:00
rcourtman 2e85079263 Hydrate Assistant finding lifecycle context 2026-05-06 19:54:04 +01:00
rcourtman 2832bd321e Invalidate stale Assistant handoff context 2026-05-06 19:36:05 +01:00
rcourtman 50fee63c9e Refresh Assistant handoff finding context 2026-05-06 19:16:57 +01:00
rcourtman 5c8070c582 Persist Assistant handoff action references 2026-05-06 18:26:21 +01:00
rcourtman aeede4fb2f Hydrate Assistant handoff resource scope 2026-05-06 17:59:55 +01:00
rcourtman 7de85ff257 Keep Patrol handoff context model-only in Assistant 2026-05-06 17:37:19 +01:00
rcourtman 87ba22b0fc Surface Patrol investigation records in Assistant context 2026-05-06 17:00:05 +01:00
rcourtman b1e179479d Retire self-hosted AI quickstart surfaces 2026-04-28 15:49:18 +01:00
rcourtman bd138beeca Improve Pulse Assistant approval continuity 2026-04-24 09:13:39 +01:00
rcourtman cc7b0980ec refactor(truenas): centralize legacy host type canonicalization 2026-03-30 13:22:50 +01:00
rcourtman 2fdf0e16c8 Support enterprise mobile approval proof backend 2026-03-29 21:52:18 +01:00
rcourtman 56c14ca19f feat(ai): add canonical truenas app config reads 2026-03-29 20:36:43 +01:00
rcourtman 298b23626b feat(ai): add canonical truenas app log reads 2026-03-29 20:13:39 +01:00
rcourtman b0ba88d541 feat(ai): add canonical truenas app control 2026-03-29 19:50:31 +01:00
rcourtman 2afb96ee13 fix(release): align api and hostagent rc contracts 2026-03-26 17:08:48 +00:00
rcourtman 6a0f564a10 fix(hosted): align tenant AI runtime with hosted lease bootstrap 2026-03-25 16:55:31 +00:00
rcourtman cc806171dc Trim dead resource graph surface 2026-03-19 14:26:30 +00:00
rcourtman 778a2577b6 feat: Pulse v6 release 2026-03-18 16:06:30 +00:00
rcourtman 95a0d7a6bd feat(backend): implement AI Patrol, Investigation, and system-wide refactors 2026-01-30 19:02:14 +00:00
rcourtman 9dcd859056 Update API handlers for AI and discovery endpoints
API layer updates:

AI Handlers:
- Better streaming response handling
- Improved error responses
- Session management improvements

Discovery Handlers:
- New discovery endpoint handlers
- Storage config handler
- Better router organization

Removed deprecated aidiscovery handlers in favor of unified approach.
2026-01-28 16:51:35 +00:00
rcourtman 7f7edfceb4 test: expand backend coverage 2026-01-25 21:08:44 +00:00
rcourtman 27f1a11acb feat: add AI Intelligence system with investigation and forecasting
Major new AI capabilities for infrastructure monitoring:

Investigation System:
- Autonomous finding investigation with configurable autonomy levels
- Investigation orchestrator with rate limiting and guardrails
- Safety checks for read-only mode enforcement
- Chat-based investigation with approval workflows

Forecasting & Remediation:
- Trend forecasting for resource capacity planning
- Remediation engine for generating fix proposals
- Circuit breaker for AI operation protection

Unified Findings:
- Unified store bridging alerts and AI findings
- Correlation and root cause analysis
- Incident coordinator with metrics recording

New Frontend:
- AI Intelligence page with patrol controls
- Investigation drawer for finding details
- Unified findings panel with actions

Supporting Infrastructure:
- Learning store for user preference tracking
- Proxmox event ingestion and correlation
- Enhanced patrol with investigation triggers
2026-01-24 22:41:43 +00:00
rcourtman f2541b0d6c Refactor: Multi-tenancy support for API and License handlers
- Updated LicenseHandlers and LicenseService to be context/tenant aware
- Refactored API router and middleware to support tenant-scoped license checks
- Updated associated tests for context-aware handlers
2026-01-22 16:42:39 +00:00
rcourtman a6a8efaa65 test: Add comprehensive test coverage across packages
New test files with expanded coverage:

API tests:
- ai_handler_test.go: AI handler unit tests with mocking
- agent_profiles_tools_test.go: Profile management tests
- alerts_endpoints_test.go: Alert API endpoint tests
- alerts_test.go: Updated for interface changes
- audit_handlers_test.go: Audit handler tests
- frontend_embed_test.go: Frontend embedding tests
- metadata_handlers_test.go, metadata_provider_test.go: Metadata tests
- notifications_test.go: Updated for interface changes
- profile_suggestions_test.go: Profile suggestion tests
- saml_service_test.go: SAML authentication tests
- sensor_proxy_gate_test.go: Sensor proxy tests
- updates_test.go: Updated for interface changes

Agent tests:
- dockeragent/signature_test.go: Docker agent signature tests
- hostagent/agent_metrics_test.go: Host agent metrics tests
- hostagent/commands_test.go: Command execution tests
- hostagent/network_helpers_test.go: Network helper tests
- hostagent/proxmox_setup_test.go: Updated setup tests
- kubernetesagent/*_test.go: Kubernetes agent tests

Core package tests:
- monitoring/kubernetes_agents_test.go, reload_test.go
- remoteconfig/client_test.go, signature_test.go
- sensors/collector_test.go
- updates/adapter_installsh_*_test.go: Install adapter tests
- updates/manager_*_test.go: Update manager tests
- websocket/hub_*_test.go: WebSocket hub tests

Library tests:
- pkg/audit/export_test.go: Audit export tests
- pkg/metrics/store_test.go: Metrics store tests
- pkg/proxmox/*_test.go: Proxmox client tests
- pkg/reporting/reporting_test.go: Reporting tests
- pkg/server/*_test.go: Server tests
- pkg/tlsutil/extra_test.go: TLS utility tests

Total: ~8000 lines of new test code
2026-01-19 19:26:18 +00:00