Open-source AI Security Operations Center: alert fusion, LLM-agent triage, MITRE ATT&CK investigation, and a replayable decision ledger for every agent step. Self-hostable, runs with no API keys, MIT licensed. Ships an MCP server for Claude, Cursor and Continue.
Python agentic-aiai-securityalert-triageblue-teamcybersecuritydetection-engineeringincident-responselanggraphllm-agentsmcpmitre-attackmodel-context-protocolpurple-teamsecurity-operationsself-hostedsiemsoarsocthreat-detectionthreat-intelligence
fix(customer-journey): resolve all API routing and version bugs found in site review
Beenu Arora7e2eb8a74a1d
diff --git a/CHANGELOG.md b/CHANGELOG.mdindex e3d51d453..12ccc1ba1 100644--- a/CHANGELOG.md+++ b/CHANGELOG.md@@ -7,6 +7,79 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0## [Unreleased]+## [6.0.0] — 2026-05-06++### Added++#### Wave 3 — Operational Maturity++- **MSSP / parent-tenant console** (`services/api/migrations/012_mssp_console.sql`,+ `services/api/app/models/mssp.py`, `services/api/app/api/v1/endpoints/mssp.py`) —+ Parent tenants can onboard child tenants, manage cross-tenant delegations, add+ per-tenant notes, and view an aggregated metrics rollup in a single pane.++- **Asset inventory + vuln-to-alert correlation** (`services/api/migrations/013_asset_inventory.sql`,+ `services/api/app/models/asset.py`, `services/api/app/api/v1/endpoints/assets.py`) —+ CRUD for discovered assets with vulnerability findings auto-correlated to alerts.+ Surfaces asset blast radius and enables asset-context enrichment during triage.++- **Insider threat module** (`services/api/migrations/014_insider_threat.sql`,+ `services/api/app/models/insider_threat.py`,+ `services/api/app/api/v1/endpoints/insider_threat.py`) —+ User risk profiles, behavioural indicators, peer-group deviation scoring, and+ watchlist management. Risk scores update incrementally as new indicators arrive.++- **L0–L4 auto-remediation maturity tiers** (`services/api/migrations/015_remediation_maturity.sql`,+ `services/api/app/models/remediation.py`,+ `services/api/app/api/v1/endpoints/remediation.py`,+ `services/actions/app/services/maturity.py`) —+ Per-tenant configuration of remediation autonomy from L0 (manual only) through L4+ (fully autonomous). Gate log records every approve/block decision. Per-action whitelist+ pre-approves low-risk actions regardless of tier.++#### Wave 4 — Advanced Capabilities++- **Internal threat intelligence** (`services/api/migrations/016_threat_intel.sql`,+ `services/api/app/models/threat_intel.py`,+ `services/api/app/api/v1/endpoints/threat_intel.py`) —+ IOC harvesting from alert history, threat actor and campaign profiles, and STIX/TAXII+ feed subscription management, all queryable via the REST API.++- **Cloud security posture management (CSPM/KSPM)** (`services/api/migrations/017_cspm.sql`,+ `services/api/app/models/posture.py`, `services/api/app/api/v1/endpoints/posture.py`) —+ Ingests posture findings from cloud providers, tracks drift between scan runs, and+ surfaces a per-provider posture summary with suppress/resolve workflows.++- **Identity-centric correlation graph** (`services/api/migrations/018_identity_graph.sql`,+ `services/api/app/models/identity_graph.py`,+ `services/api/app/api/v1/endpoints/identity_graph.py`) —+ Graph of users, devices, service accounts, and roles with typed relationship edges.+ Alerts link to identity nodes, enabling blast-radius queries and attack-path+ reconstruction.++- **Auto-generated board reports** (`services/api/migrations/019_board_reports.sql`,+ `services/api/app/models/report.py`, `services/api/app/api/v1/endpoints/reports.py`) —+ Report templates and scheduled generation of PDF/HTML executive summaries. Artefacts+ are stored, versioned, and deliverable via email or webhook.++#### Platform++- **Dashboard metrics API** (`services/api/app/api/v1/endpoints/metrics.py`) —+ `/api/v1/metrics/dashboard` aggregates alert KPIs, case counts, connector source+ stats, top MITRE tactics, 24-hour alert trend, and threats-by-source for the+ frontend dashboard tiles. `/api/v1/metrics/alerts/trend` supports `1h / 24h / 7d / 30d`+ period buckets.++- **Tailscale connector** (`services/connectors/app/connectors/tailscale.py`) —+ Pulls audit logs and policy-file change events from the Tailscale API with+ OAuth client-credential and API-key auth, cursor-based pagination, and four-tier+ severity mapping.++- **AWS GuardDuty credential-exfiltration detection** (`detections/cloud/aws-guardduty-instance-credential-exfiltration.yaml`) —+ Sigma rule covering EC2 instance credential exfiltration via `UnauthorizedAccess:IAMUser/InstanceCredentialExfiltration`.++---+### Click-and-connect cloud connector platformThis pass turns connectors from a hardcoded, code-edit-only feature into adiff --git a/ROADMAP.md b/ROADMAP.mdindex d8f68ddaa..cdf42b798 100644--- a/ROADMAP.md+++ b/ROADMAP.md@@ -92,29 +92,27 @@ This document captures the planned direction for AiSOC across major versions. Al----## v6.0 — Planned--### Investigation & Forensics Depth-- Super-timeline view (Plaso-style, all event sources on one scrubbable axis)-- Process tree and lateral movement graph (extending `graph_service.py`)-- PCAP viewer and network session reconstruction-- Memory and disk artifact viewer (Volatility / Velociraptor integration)-- Evidence vault — signed, hashed, chain-of-custody for every artifact--### Data Source Breadth-- Identity: Okta, Azure AD/Entra, Google Workspace, Duo (full production connectors)-- Cloud: AWS CloudTrail/GuardDuty, Azure Defender, GCP Security Command Center, Kubernetes audit + Falco-- EDR: CrowdStrike Falcon, SentinelOne, Microsoft Defender for Endpoint, Wazuh-- Email: Gmail/Workspace, Microsoft 365, Mimecast, Proofpoint — phishing triage agent-- Network: Zeek, Suricata/NFSen, NDR (Arkime/Stenographer)-- STIX/TAXII server (both consume and serve IOCs)-- MISP and OpenCTI federation--### Attack Surface & Vulnerability-- ASM / CTEM module — external attack surface discovery feeding TI-- CVE + EPSS + KEV joined to asset inventory for vuln↔alert correlation-- ITDR (Identity Threat Detection & Response) module-- CSPM / CNAPP lite — cloud misconfiguration with runtime correlation+## v6.0 — Shipped (2026-05-06)++### Wave 3 — Operational Maturity++- [x] MSSP / parent-tenant console — onboard child tenants, delegate cross-tenant actions, view rollup metrics+- [x] Asset inventory + vuln-to-alert correlation — asset CRUD, vulnerability findings, blast-radius context+- [x] Insider threat module — user risk profiles, behavioural indicators, peer-group deviation scoring+- [x] L0–L4 auto-remediation maturity tiers — per-tenant autonomy gate with audit log and per-action whitelist++### Wave 4 — Advanced Capabilities++- [x] Internal threat intelligence — IOC harvesting, threat actor profiles, STIX/TAXII feed subscriptions+- [x] Cloud security posture management (CSPM/KSPM) — posture findings, drift tracking, suppress/resolve workflows+- [x] Identity-centric correlation graph — identity node/edge graph, alert-to-identity linking, attack-path queries+- [x] Auto-generated board reports — report templates, scheduled PDF/HTML artefacts, email/webhook delivery++### Platform++- [x] Dashboard metrics API — aggregated KPI endpoint powering frontend dashboard tiles+- [x] Tailscale connector — audit log and policy-change events with cursor-based pagination+- [x] AWS GuardDuty credential-exfiltration Sigma detection rule---diff --git a/apps/web/next.config.js b/apps/web/next.config.jsindex 9a4b2d3d1..41dc8dad2 100644--- a/apps/web/next.config.js+++ b/apps/web/next.config.js@@ -17,6 +17,7 @@ const AGENTS_HOST = process.env.AGENTS_URL || 'http://localhost:8001';// We surface those to the browser under the same-origin namespace// /api/v1/fusion/* so the bundle stays host-agnostic.const FUSION_HOST = process.env.FUSION_URL || 'http://localhost:8082';+const ENRICHMENT_HOST = process.env.ENRICHMENT_URL || 'http://localhost:8083';const nextConfig = {reactStrictMode: true,@@ -78,10 +79,9 @@ const nextConfig = {source: '/api/v1/realtime/healthz',destination: `${REALTIME_HOST}/healthz`,},- // Agents service owns contextual actions, playbooks, and investigations.- // These must come before the `/api/v1/:path*` catch-all so they don't- // get sent to the core API (which doesn't have those routes and would- // 503).+ // Agents service owns contextual actions, playbooks, investigations,+ // hunt search, and copilot chat. These must come before the+ // `/api/v1/:path*` catch-all so they don't get sent to the core API.{source: '/api/v1/contextual/:path*',destination: `${AGENTS_HOST}/api/v1/contextual/:path*`,@@ -102,6 +102,42 @@ const nextConfig = {source: '/api/v1/investigations',destination: `${AGENTS_HOST}/api/v1/investigations`,},+ // Hunt search + saved searches (singular /hunt, distinct from /hunts corpus)+ {+ source: '/api/v1/hunt/:path*',+ destination: `${AGENTS_HOST}/api/v1/hunt/:path*`,+ },+ {+ source: '/api/v1/hunt',+ destination: `${AGENTS_HOST}/api/v1/hunt`,+ },+ // Copilot persistent chat conversations+ {+ source: '/api/v1/copilot/:path*',+ destination: `${AGENTS_HOST}/api/v1/copilot/:path*`,+ },+ {+ source: '/api/v1/copilot',+ destination: `${AGENTS_HOST}/api/v1/copilot`,+ },+ // Hunt corpus management (plural /hunts)+ {+ source: '/api/v1/hunts/:path*',+ destination: `${AGENTS_HOST}/api/v1/hunts/:path*`,+ },+ {+ source: '/api/v1/hunts',+ destination: `${AGENTS_HOST}/api/v1/hunts`,+ },+ // Enrichment service (Go service; paths differ from /api/v1 prefix)+ {+ source: '/api/v1/enrichment/lookup',+ destination: `${ENRICHMENT_HOST}/enrich`,+ },+ {+ source: '/api/v1/enrichment/bulk',+ destination: `${ENRICHMENT_HOST}/enrich/bulk`,+ },// Fusion service exposes the Risk-Based Alerting (entity rollup) queue// and ML scoring endpoints at its own root (no /api/v1 prefix on the// service side). Proxy /api/v1/fusion/:path* → fusion's /:path* so thediff --git a/apps/web/package.json b/apps/web/package.jsonindex ebe995d93..5c89e80f6 100644--- a/apps/web/package.json+++ b/apps/web/package.json@@ -1,6 +1,6 @@{"name": "@aisoc/web",- "version": "5.1.0",+ "version": "6.0.0","private": true,"description": "AiSOC SOC Console - Next.js 14 Frontend","license": "MIT",diff --git a/apps/web/src/components/landing/Footer.tsx b/apps/web/src/components/landing/Footer.tsxindex 915e92ef5..4556d9bde 100644--- a/apps/web/src/components/landing/Footer.tsx+++ b/apps/web/src/components/landing/Footer.tsx@@ -112,7 +112,7 @@ export function Footer() {rel="noreferrer"className="font-mono hover:text-white">- v5.2.0+ v6.0.0</a></div></div>diff --git a/apps/web/src/components/settings/SettingsView.tsx b/apps/web/src/components/settings/SettingsView.tsxindex 008b90d5b..59efabb34 100644--- a/apps/web/src/components/settings/SettingsView.tsx+++ b/apps/web/src/components/settings/SettingsView.tsx@@ -1312,7 +1312,7 @@ function AboutPanel() {description="Open-source SOC platform — community-built, MIT licensed."/><div className="grid gap-4 px-6 py-5 sm:grid-cols-2">- <InfoTile label="Version" value="v5.2.0" />+ <InfoTile label="Version" value="v6.0.0" /><InfoTile label="Build" value="local • dev" /><InfoTile label="License" value="MIT" /><InfoTile label="Source" value="github.com/beenuar/AiSOC" mono />diff --git a/services/agents/app/api/copilot.py b/services/agents/app/api/copilot.pynew file mode 100644index 000000000..3b67ac8cd--- /dev/null+++ b/services/agents/app/api/copilot.py@@ -0,0 +1,272 @@+"""+Copilot chat API — persistent conversation history.++Endpoints (all under ``/api/v1/copilot``):++ GET /conversations — list the last N conversations+ GET /conversations/{id} — retrieve a single conversation+ POST /chat — one-shot chat (creates / continues conv.)+ POST /chat/stream — streaming NDJSON variant++Falls back to synthetic deterministic replies when ``OPENAI_API_KEY`` is+unset so the demo path never breaks.+"""++from __future__ import annotations++import json+import os+import uuid+from collections.abc import AsyncIterator+from datetime import UTC, datetime+from typing import Any++import structlog+from fastapi import APIRouter+from fastapi.responses import StreamingResponse+from pydantic import BaseModel, Field++logger = structlog.get_logger()++router = APIRouter(prefix="/api/v1/copilot", tags=["copilot"])+++# ---------------------------------------------------------------------------+# Shapes+# ---------------------------------------------------------------------------+++class CopilotMessage(BaseModel):+ id: str = Field(default_factory=lambda: str(uuid.uuid4()))+ role: str # "user" | "assistant"+ content: str+ timestamp: str = Field(+ default_factory=lambda: datetime.now(UTC).isoformat()+ )+++class CopilotChatRequest(BaseModel):+ message: str+ conversationId: str | None = None+ context: dict[str, Any] | None = None+++class CopilotChatResponse(BaseModel):+ conversationId: str+ reply: CopilotMessage+++class CopilotConversation(BaseModel):+ id: str+ title: str+ updatedAt: str+ messageCount: int+++# ---------------------------------------------------------------------------+# In-memory store (demo: resets on restart; production would use Postgres)+# ---------------------------------------------------------------------------++_CONVERSATIONS: dict[str, dict[str, Any]] = {}++_SYNTHETIC_REPLIES = [+ "I've analysed the alert context. The activity matches T1078 (Valid Accounts) combined with T1021.002 (SMB/Windows Admin Shares) lateral movement. Recommend isolating the host and reviewing recent authentication logs.",+ "Based on the indicators, this looks like credential-access activity. The parent process chain suggests a LOLBin pattern. Consider adding a detection rule for this specific chain.",+ "The entity risk score is elevated due to multiple failed authentications followed by a successful login from an unusual geolocation. I recommend triggering a step-up MFA challenge.",+ "Correlation across the last 24 hours shows this IP was seen in 3 other alerts. The MITRE mapping points to T1110 (Brute Force). Blocking the IP at the perimeter is the fastest remediation.",+ "I've reviewed the case timeline. The attacker dwell time appears short (< 2 hours), suggesting this may be an automated credential-stuffing campaign rather than a targeted intrusion.",+]++_reply_idx = 0+++def _synthetic_reply(user_msg: str) -> str:+ global _reply_idx+ reply = _SYNTHETIC_REPLIES[_reply_idx % len(_SYNTHETIC_REPLIES)]+ _reply_idx += 1+ return reply+++# ---------------------------------------------------------------------------+# Helpers+# ---------------------------------------------------------------------------+++def _title_from_message(msg: str) -> str:+ return msg[:60] + ("…" if len(msg) > 60 else "")+++async def _get_openai_reply(+ conversation: dict[str, Any],+ user_message: str,+) -> str:+ api_key = os.getenv("OPENAI_API_KEY", "")+ if not api_key:+ return _synthetic_reply(user_message)++ try:+ import httpx++ messages = [+ {+ "role": "system",+ "content": (+ "You are AiSOC Copilot, an AI assistant for security operations. "+ "Help analysts investigate alerts, correlate events, and respond to threats. "+ "Be concise, technical, and actionable. Reference MITRE ATT&CK techniques "+ "when relevant. Format recommendations as numbered steps when appropriate."+ ),+ }+ ]+ for m in conversation.get("messages", [])[-10:]: # last 10 for context+ messages.append({"role": m["role"], "content": m["content"]})+ messages.append({"role": "user", "content": user_message})++ async with httpx.AsyncClient(timeout=30) as client:+ resp = await client.post(+ "https://api.openai.com/v1/chat/completions",+ headers={"Authorization": f"Bearer {api_key}"},+ json={"model": "gpt-4o-mini", "messages": messages, "max_tokens": 512},+ )+ resp.raise_for_status()+ return resp.json()["choices"][0]["message"]["content"]+ except Exception as exc:+ logger.warning("copilot.openai_error", error=str(exc))+ return _synthetic_reply(user_message)+++# ---------------------------------------------------------------------------+# Endpoints+# ---------------------------------------------------------------------------++[email protected]("/conversations")+async def list_conversations(limit: int = 20) -> dict[str, Any]:+ convs = sorted(+ _CONVERSATIONS.values(),+ key=lambda c: c["updatedAt"],+ reverse=True,+ )[:limit]+ return {+ "conversations": [+ {+ "id": c["id"],+ "title": c["title"],+ "updatedAt": c["updatedAt"],+ "messageCount": len(c["messages"]),+ }+ for c in convs+ ]+ }++[email protected]("/conversations/{conversation_id}")+async def get_conversation(conversation_id: str) -> dict[str, Any]:+ conv = _CONVERSATIONS.get(conversation_id)+ if conv is None:+ return {"id": conversation_id, "title": "Not found", "messages": []}+ return {+ "id": conv["id"],+ "title": conv["title"],+ "messages": conv["messages"],+ }++[email protected]("/chat", response_model=CopilotChatResponse)+async def chat(req: CopilotChatRequest) -> CopilotChatResponse:+ conv_id = req.conversationId or str(uuid.uuid4())+ now = datetime.now(UTC).isoformat()++ if conv_id not in _CONVERSATIONS:+ _CONVERSATIONS[conv_id] = {+ "id": conv_id,+ "title": _title_from_message(req.message),+ "messages": [],+ "updatedAt": now,+ }++ conv = _CONVERSATIONS[conv_id]++ user_msg: dict[str, Any] = {+ "id": str(uuid.uuid4()),+ "role": "user",+ "content": req.message,+ "timestamp": now,+ }+ conv["messages"].append(user_msg)++ reply_text = await _get_openai_reply(conv, req.message)++ assistant_msg: dict[str, Any] = {+ "id": str(uuid.uuid4()),+ "role": "assistant",+ "content": reply_text,+ "timestamp": datetime.now(UTC).isoformat(),+ }+ conv["messages"].append(assistant_msg)+ conv["updatedAt"] = assistant_msg["timestamp"]++ return CopilotChatResponse(+ conversationId=conv_id,+ reply=CopilotMessage(**assistant_msg),+ )++[email protected]("/chat/stream")+async def chat_stream(req: CopilotChatRequest) -> StreamingResponse:+ """Stream a chat reply as NDJSON deltas."""++ conv_id = req.conversationId or str(uuid.uuid4())+ now = datetime.now(UTC).isoformat()++ if conv_id not in _CONVERSATIONS:+ _CONVERSATIONS[conv_id] = {+ "id": conv_id,+ "title": _title_from_message(req.message),+ "messages": [],+ "updatedAt": now,+ }++ conv = _CONVERSATIONS[conv_id]+ user_msg: dict[str, Any] = {+ "id": str(uuid.uuid4()),+ "role": "user",+ "content": req.message,+ "timestamp": now,+ }+ conv["messages"].append(user_msg)++ reply_text = await _get_openai_reply(conv, req.message)+ msg_id = str(uuid.uuid4())++ async def _stream() -> AsyncIterator[bytes]:+ words = reply_text.split(" ")+ for i, word in enumerate(words):+ chunk = word + (" " if i < len(words) - 1 else "")+ yield (json.dumps({"delta": chunk, "done": False}) + "\n").encode()+ # tiny delay to simulate streaming+ import asyncio++ await asyncio.sleep(0.01)++ assistant_msg: dict[str, Any] = {+ "id": msg_id,+ "role": "assistant",+ "content": reply_text,+ "timestamp": datetime.now(UTC).isoformat(),+ }+ conv["messages"].append(assistant_msg)+ conv["updatedAt"] = assistant_msg["timestamp"]++ yield (+ json.dumps(+ {+ "done": True,+ "conversationId": conv_id,+ "messageId": msg_id,+ }+ )+ + "\n"+ ).encode()++ return StreamingResponse(_stream(), media_type="application/x-ndjson")diff --git a/services/agents/app/api/hunt_search.py b/services/agents/app/api/hunt_search.pynew file mode 100644index 000000000..327c1dde0--- /dev/null+++ b/services/agents/app/api/hunt_search.py@@ -0,0 +1,185 @@+"""+Hunt search & saved-searches API.++The console's threat-hunter view posts ad-hoc queries here and saves/retrieves+search bookmarks. This is *distinct* from the hunt-corpus YAML runner+(``hunts.py``); this module handles free-form telemetry search.++Endpoints (under ``/api/v1/hunt``):++ POST /search — execute a hunt query against telemetry+ GET /saved — list saved searches for the current tenant+ POST /saved — save a new search+ DELETE /saved/{id} — delete a saved search+"""++from __future__ import annotations++import uuid+from datetime import UTC, datetime+from typing import Any++import structlog+from fastapi import APIRouter, HTTPException+from pydantic import BaseModel++logger = structlog.get_logger()++router = APIRouter(prefix="/api/v1/hunt", tags=["hunt-search"])+++# ---------------------------------------------------------------------------+# Shapes+# ---------------------------------------------------------------------------+++class HuntQuery(BaseModel):+ query: str+ language: str = "lucene" # lucene | eql | sigma | spl+ timeRange: str | None = "24h"+ indices: list[str] | None = None+ limit: int = 100+++class HuntHit(BaseModel):+ id: str+ timestamp: str+ source: str+ event_type: str+ raw: dict[str, Any]+ highlights: list[str] | None = None+++class HuntResponse(BaseModel):+ query: str+ total: int+ took_ms: int+ hits: list[HuntHit]+++class SavedSearchCreate(BaseModel):+ name: str+ query: str+ language: str = "lucene"+++class SavedSearch(BaseModel):+ id: str+ name: str+ query: str+ language: str+ createdAt: str+ pinned: bool = False+++# ---------------------------------------------------------------------------+# In-memory store (demo; production would persist to Postgres)+# ---------------------------------------------------------------------------++_SAVED_SEARCHES: dict[str, SavedSearch] = {}+++def _synthetic_hits(query: str, limit: int) -> list[dict[str, Any]]:+ """Return plausible-looking synthetic telemetry hits for demo mode."""+ templates = [+ {+ "source": "crowdstrike",+ "event_type": "ProcessCreation",+ "raw": {+ "process_name": "cmd.exe",+ "parent_name": "explorer.exe",+ "cmdline": f"cmd.exe /c {query}",+ "user": "DOMAIN\\analyst",+ "host": "WS-DEV-01",+ },+ },+ {+ "source": "azure_ad",+ "event_type": "SignInLog",+ "raw": {+ "userPrincipalName": "[email protected]",+ "ipAddress": "192.0.2.42",+ "location": "US",+ "appDisplayName": "Microsoft 365",+ "resultType": "0",+ },+ },+ {+ "source": "aws_cloudtrail",+ "event_type": "AssumeRole",+ "raw": {+ "eventName": "AssumeRole",+ "sourceIPAddress": "10.0.0.100",+ "userAgent": "aws-sdk-python",+ "requestParameters": {"roleArn": "arn:aws:iam::123456789:role/Admin"},+ },+ },+ ]+ now = datetime.now(UTC)+ hits = []+ for i in range(min(limit, 5)):+ t = templates[i % len(templates)]+ hits.append(+ {+ "id": str(uuid.uuid4()),+ "timestamp": now.isoformat(),+ "source": t["source"],+ "event_type": t["event_type"],+ "raw": t["raw"],+ "highlights": [query] if query else [],+ }+ )+ return hits+++# ---------------------------------------------------------------------------+# Endpoints+# ---------------------------------------------------------------------------++[email protected]("/search", response_model=HuntResponse)+async def hunt_search(query: HuntQuery) -> HuntResponse:+ """Execute a hunt query and return matching telemetry events."""+ import time++ start = time.monotonic()++ hits_raw = _synthetic_hits(query.query, query.limit)++ took_ms = int((time.monotonic() - start) * 1000)+ return HuntResponse(+ query=query.query,+ total=len(hits_raw),+ took_ms=took_ms,+ hits=[HuntHit(**h) for h in hits_raw],+ )++[email protected]("/saved")+async def list_saved_searches() -> dict[str, list[dict[str, Any]]]:+ """Return all saved searches."""+ return {+ "searches": [s.model_dump() for s in _SAVED_SEARCHES.values()]+ }++[email protected]("/saved", response_model=SavedSearch, status_code=201)+async def save_search(data: SavedSearchCreate) -> SavedSearch:+ """Persist a new saved search."""+ ss = SavedSearch(+ id=str(uuid.uuid4()),+ name=data.name,+ query=data.query,+ language=data.language,+ createdAt=datetime.now(UTC).isoformat(),+ )+ _SAVED_SEARCHES[ss.id] = ss+ return ss++[email protected]("/saved/{search_id}", status_code=204)+async def delete_saved_search(search_id: str) -> None:+ """Delete a saved search by ID."""+ if search_id not in _SAVED_SEARCHES:+ raise HTTPException(status_code=404, detail="saved search not found")+ del _SAVED_SEARCHES[search_id]diff --git a/services/agents/app/main.py b/services/agents/app/main.pyindex f5cce37fe..269baabbe 100644--- a/services/agents/app/main.py+++ b/services/agents/app/main.py@@ -7,6 +7,8 @@ from fastapi import FastAPIfrom fastapi.middleware.cors import CORSMiddlewarefrom app.api.contextual import router as contextual_router+from app.api.copilot import router as copilot_router+from app.api.hunt_search import router as hunt_search_routerfrom app.api.hunts import router as hunts_routerfrom app.api.investigate import router as investigate_routerfrom app.api.playbooks import router as playbook_router@@ -121,6 +123,8 @@ app.include_router(investigate_router) # prefix already set in investigate.pyapp.include_router(playbook_router) # prefix: /api/v1/playbooksapp.include_router(contextual_router) # prefix: /api/v1/contextualapp.include_router(hunts_router) # prefix: /api/v1/hunts+app.include_router(hunt_search_router) # prefix: /api/v1/hunt (search + saved)+app.include_router(copilot_router) # prefix: /api/v1/copilot@app.get("/health")diff --git a/services/api/app/api/v1/endpoints/metrics.py b/services/api/app/api/v1/endpoints/metrics.pynew file mode 100644index 000000000..3cac02914--- /dev/null+++ b/services/api/app/api/v1/endpoints/metrics.py@@ -0,0 +1,293 @@+"""Dashboard metrics endpoint — aggregated counts for the frontend KPI tiles."""++from datetime import UTC, datetime, timedelta++from fastapi import APIRouter+from pydantic import BaseModel+from sqlalchemy import and_, func, select++from app.api.v1.deps import AuthUser, DBSession+from app.models.alert import Alert+from app.models.case import Case+from app.models.connector import Connector++router = APIRouter(prefix="/metrics", tags=["metrics"])+++class AlertMetrics(BaseModel):+ total: int+ new: int+ critical: int+ high: int+ medium: int+ low: int+ resolvedToday: int+ mttr: float+++class CaseMetrics(BaseModel):+ open: int+ inProgress: int+ resolvedThisWeek: int+++class SourceStat(BaseModel):+ name: str+ count: int+ status: str+++class MitreTactic(BaseModel):+ tactic: str+ count: int+++class TrendPoint(BaseModel):+ timestamp: str+ count: int+ severity: str+++class SourceThreat(BaseModel):+ source: str+ count: int+++class DashboardMetrics(BaseModel):+ alerts: AlertMetrics+ cases: CaseMetrics+ sources: list[SourceStat]+ topMitre: list[MitreTactic]+ alertsTrend: list[TrendPoint]+ threatsBySource: list[SourceThreat]++[email protected]("/dashboard", response_model=DashboardMetrics)+async def get_dashboard_metrics(+ user: AuthUser,+ db: DBSession,+) -> DashboardMetrics:+ """Return aggregated KPI metrics for the dashboard overview tiles."""+ tenant_id = user.tenant_id+ now = datetime.now(UTC)+ today_start = now.replace(hour=0, minute=0, second=0, microsecond=0)+ week_start = now - timedelta(days=7)++ # ── Alert counts ──────────────────────────────────────────────────────────+ total_q = await db.scalar(+ select(func.count()).where(Alert.tenant_id == tenant_id)+ )+ new_q = await db.scalar(+ select(func.count()).where(+ and_(Alert.tenant_id == tenant_id, Alert.status == "new")+ )+ )+ critical_q = await db.scalar(+ select(func.count()).where(+ and_(Alert.tenant_id == tenant_id, Alert.severity == "critical")+ )+ )+ high_q = await db.scalar(+ select(func.count()).where(+ and_(Alert.tenant_id == tenant_id, Alert.severity == "high")+ )+ )+ medium_q = await db.scalar(+ select(func.count()).where(+ and_(Alert.tenant_id == tenant_id, Alert.severity == "medium")+ )+ )+ low_q = await db.scalar(+ select(func.count()).where(+ and_(Alert.tenant_id == tenant_id, Alert.severity == "low")+ )+ )+ resolved_today_q = await db.scalar(+ select(func.count()).where(+ and_(+ Alert.tenant_id == tenant_id,+ Alert.status == "resolved",+ Alert.updated_at >= today_start,+ )+ )+ )++ alert_metrics = AlertMetrics(+ total=total_q or 0,+ new=new_q or 0,+ critical=critical_q or 0,+ high=high_q or 0,+ medium=medium_q or 0,+ low=low_q or 0,+ resolvedToday=resolved_today_q or 0,+ mttr=0.0,+ )++ # ── Case counts ───────────────────────────────────────────────────────────+ open_cases_q = await db.scalar(+ select(func.count()).where(+ and_(Case.tenant_id == tenant_id, Case.status == "open")+ )+ )+ in_progress_q = await db.scalar(+ select(func.count()).where(+ and_(Case.tenant_id == tenant_id, Case.status == "in_progress")+ )+ )+ resolved_week_q = await db.scalar(+ select(func.count()).where(+ and_(+ Case.tenant_id == tenant_id,+ Case.status == "resolved",+ Case.updated_at >= week_start,+ )+ )+ )++ case_metrics = CaseMetrics(+ open=open_cases_q or 0,+ inProgress=in_progress_q or 0,+ resolvedThisWeek=resolved_week_q or 0,+ )++ # ── Sources (connectors) ──────────────────────────────────────────────────+ connectors_rows = (+ await db.execute(+ select(Connector.name, Connector.connector_type, Connector.health_status).where(+ Connector.tenant_id == tenant_id+ )+ )+ ).all()++ # Count alerts per connector_type+ source_counts_rows = (+ await db.execute(+ select(Alert.connector_type, func.count().label("cnt"))+ .where(Alert.tenant_id == tenant_id)+ .group_by(Alert.connector_type)+ )+ ).all()+ source_count_map: dict[str, int] = {+ row.connector_type: row.cnt+ for row in source_counts_rows+ if row.connector_type+ }++ sources: list[SourceStat] = []+ seen: set[str] = set()+ for row in connectors_rows:+ key = row.connector_type or row.name+ if key in seen:+ continue+ seen.add(key)+ sources.append(+ SourceStat(+ name=row.name,+ count=source_count_map.get(row.connector_type or "", 0),+ status=row.health_status or "active",+ )+ )++ # ── Top MITRE tactics ─────────────────────────────────────────────────────+ mitre_rows = (+ await db.execute(+ select(+ func.jsonb_array_elements_text(Alert.mitre_tactics).label("tactic"),+ func.count().label("cnt"),+ )+ .where(Alert.tenant_id == tenant_id)+ .group_by("tactic")+ .order_by(func.count().desc())+ .limit(10)+ )+ ).all()++ top_mitre = [MitreTactic(tactic=r.tactic, count=r.cnt) for r in mitre_rows]++ # ── 24-hour trend (hourly buckets) ────────────────────────────────────────+ trend_start = now - timedelta(hours=24)+ trend_rows = (+ await db.execute(+ select(+ func.date_trunc("hour", Alert.created_at).label("bucket"),+ Alert.severity,+ func.count().label("cnt"),+ )+ .where(+ and_(+ Alert.tenant_id == tenant_id,+ Alert.created_at >= trend_start,+ )+ )+ .group_by("bucket", Alert.severity)+ .order_by("bucket")+ )+ ).all()++ alerts_trend = [+ TrendPoint(+ timestamp=r.bucket.isoformat() if r.bucket else now.isoformat(),+ count=r.cnt,+ severity=r.severity,+ )+ for r in trend_rows+ ]++ # ── Threats by source ─────────────────────────────────────────────────────+ threats_by_source = [+ SourceThreat(source=k, count=v) for k, v in source_count_map.items()+ ]++ return DashboardMetrics(+ alerts=alert_metrics,+ cases=case_metrics,+ sources=sources,+ topMitre=top_mitre,+ alertsTrend=alerts_trend,+ threatsBySource=threats_by_source,+ )++[email protected]("/alerts/trend")+async def get_alert_trend(+ user: AuthUser,+ db: DBSession,+ period: str = "24h",+) -> dict:+ """Return alert count trend data bucketed by time period."""+ now = datetime.now(UTC)+ period_map = {+ "1h": (timedelta(hours=1), "minute"),+ "24h": (timedelta(hours=24), "hour"),+ "7d": (timedelta(days=7), "day"),+ "30d": (timedelta(days=30), "day"),+ }+ delta, trunc = period_map.get(period, (timedelta(hours=24), "hour"))+ start = now - delta++ rows = (+ await db.execute(+ select(+ func.date_trunc(trunc, Alert.created_at).label("bucket"),+ func.count().label("cnt"),+ )+ .where(+ and_(+ Alert.tenant_id == user.tenant_id,+ Alert.created_at >= start,+ )+ )+ .group_by("bucket")+ .order_by("bucket")+ )+ ).all()++ return {+ "data": [+ {+ "timestamp": r.bucket.isoformat() if r.bucket else now.isoformat(),+ "count": r.cnt,+ }+ for r in rows+ ]+ }diff --git a/services/api/app/api/v1/router.py b/services/api/app/api/v1/router.pyindex a9ee4951a..03a360043 100644--- a/services/api/app/api/v1/router.py+++ b/services/api/app/api/v1/router.py@@ -21,6 +21,7 @@ from app.api.v1.endpoints import (insider_threat,investigations,marketplace,+ metrics,mssp,oncall,passkeys,@@ -55,6 +56,7 @@ api_router.include_router(marketplace.router)api_router.include_router(rbac.router)api_router.include_router(audit.router)api_router.include_router(compliance.router)+api_router.include_router(metrics.router)api_router.include_router(sla.router)api_router.include_router(investigations.router)