Self-hosted homelab log intelligence over MCP, CLI, and REST with SQLite/FTS.
It collects logs and operational evidence, stores them in SQLite with FTS5 search, and exposes one shared intelligence layer through CLI, REST, MCP, and a bundled browser workspace.
Cortex began as a syslog receiver. It now covers network logs, Docker, managed files, OpenTelemetry logs, metrics, and traces, host heartbeats, fleet inventory, shell and agent activity, and Claude, Codex, Gemini CLI, and Antigravity transcripts. It correlates those sources into timelines, incidents, and an evidence-backed topology graph without making the graph a second source of truth.
At a glance
Area
What Cortex provides
Ingest
UDP/TCP syslog, OTLP/HTTP logs, metrics, and traces, Docker logs and events, managed file tails, host heartbeats, AI transcripts, shell history, agent command records, and fleet inventory
Storage
SQLite in WAL mode, FTS5 full-text search, bounded metadata, retention, storage budgets, maintenance jobs, checkpoints, and 63 sequential schema migrations
SSH and API inventory collectors, host state, service topology, container and route relationships, redacted evidence, and rebuildable graph projections
AI operations
Claude, Codex, Gemini CLI, and Antigravity session indexing; skill, MCP, and hook event extraction where each provider exposes them; incident clustering; and guarded local LLM assessments
Interfaces
Native CLI, one action-dispatched MCP tool, authenticated REST APIs, MCP prompts and resources, an MCP Apps search widget, and a bundled investigation workspace
Operations
Setup and repair, diagnostics, Compose control, backup, integrity checks, WAL checkpoints, vacuum, update workflows, agents, and health endpoints
[!IMPORTANT]
Cortex is designed for a trusted homelab or small private fleet. It is not a clustered log warehouse, a general-purpose SIEM, or a safe place to expose unauthenticated administrative surfaces to the public internet.
Start a fresh agent task and invoke $install-cortex. It chooses server or client-only role, delegates to the canonical installer/setup engine, configures auth/network/storage, and verifies the live client connection.
The npm launcher remains the fastest manual path for local CLI and stdio MCP use:
The launcher requires Node.js 18 or newer. It downloads a checksum-verified native release binary for Linux x64, Linux ARM64, macOS ARM64, or Windows x64.
Build from source with the current stable Rust toolchain:
git clone https://github.com/dinglebear-ai/cortex.git
cd cortex
mise install # optional, but pins the repository tools
just build
./.cache/cargo/debug/cortex --version
Start a local server
The full daemon starts UDP and TCP syslog receivers plus the shared HTTP server. Use separate MCP and REST tokens:
For a managed local deployment, cortex setup repair creates or repairs the Cortex home, Compose assets, data paths, and missing 64-character MCP and REST tokens without replacing existing token values.
Connect an MCP client
Query-only stdio mode reads the configured local database and starts no network listeners:
Then narrow the investigation with tail, errors, search, timeline, or context before using broader analysis operations.
How Cortex is built
Cortex is one Rust binary with multiple operating modes. The same application and service layer backs the CLI, REST handlers, and MCP handlers, so validation, limits, identity resolution, redaction, and business rules do not belong to one transport alone.
The daemon supervises its receivers and background services with cooperative cancellation. Shutdown drains HTTP requests, gives maintenance tasks 10 seconds to finish before abort-and-join, gives ingest 5 seconds to flush, and then attempts a WAL checkpoint. Already-running blocking SQLite calls cannot be cancelled by Tokio.
Background services include:
Retention and storage-budget enforcement
WAL and FTS maintenance
Docker ingest supervision
File-tail supervision
Error-signature scanning
Notification evaluation, dispatch, and digest scheduling
Inventory refresh and backfill
Graph projection refresh
AI-session and timeline rollups
Database optimization and maintenance jobs
Heavy analytical reads and maintenance jobs have separate concurrency controls so one expensive investigation cannot starve the ingest path.
Ingestion
All log-like sources are normalized into the same durable log model, enriched where safe, scrubbed where configured, and written through bounded batch paths.
Syslog over UDP and TCP
Cortex listens on the same configurable port for UDP and TCP syslog. It parses common RFC 3164 and RFC 5424 shapes, preserves the raw frame, records sender identity, normalizes severity and facility, and enriches known application formats.
Relevant defaults:
Bind: 0.0.0.0:1514
Maximum message: 8 KiB
Maximum concurrent TCP connections: 512
TCP idle timeout: 300 seconds
Writer batch: 100 records or 500 ms
Write queue capacity: 10,000 records
Syslog has no application-layer authentication. Restrict senders with network controls and CORTEX_ALLOWED_SOURCE_CIDRS when the listener is reachable beyond a trusted network.
Built-in enrichment recognizes useful signals from AdGuard, Authelia, Docker lifecycle events, fail2ban, Linux kernel and OOM events, SWAG, reverse-proxy logs, and host-local Cortex Docker agent metadata. Source gates can restrict enrichment that would otherwise trust a marker inside an unauthenticated syslog body.
OpenTelemetry ingestion
Cortex accepts protobuf OTLP/HTTP export requests on the shared HTTP listener. Logs use POST /v1/logs with a 4 MiB body limit and flow through the normal writer. Metrics use POST /v1/metrics and traces use POST /v1/traces, each with an 8 MiB body limit and dedicated storage. Metric requests admit at most 5,000 points and also enforce a bounded normalized payload budget.
Current OTLP scope is intentionally narrow:
Logs, traces, and metrics over HTTP/protobuf are supported.
OTLP/gRPC is not implemented.
All three OTLP endpoints authenticate with CORTEX_TOKEN — the same static MCP bearer token that guards POST /mcp, read from the managed ~/.cortex/.env on a deployed host. It is notCORTEX_API_TOKEN (REST /api/*) and notCORTEX_API_ADMIN_TOKEN. Loopback and trusted-gateway policies skip the check. An OAuth-only deployment with no static token denies OTLP outright, because machine exporters have no OAuth flow — so a non-loopback OAuth-only OTLP exposure is rejected at startup unless CORTEX_TOKEN is set.
Docker logs and events
Cortex supports two Docker collection paths:
Host-local agent, the preferred multi-host path. The host-local cortex agent reads the local Docker socket, converts logs and lifecycle events into bounded records, and forwards them to the server without changing Docker's daemon logging driver.
Central pull compatibility mode, an optional server-side collector for explicitly configured Docker Engine or docker-socket-proxy HTTP endpoints. It records per-container checkpoints and reconnects with bounded exponential backoff.
Central pull is disabled by default. The CORTEX_DOCKER_HOSTS shorthand expands hosts into insecure http://host:2375 endpoints and should only be used on a tightly controlled private network. A hosts file supports explicit base URLs and safer endpoint configuration.
Managed file tails
Managed file-tail sources are persisted in a registry and supervised by the daemon. Add, remove, list, and inspect sources through the CLI, REST, or the file_tails MCP admin action.
The path policy rejects unsafe targets, including paths outside configured roots, symlink escapes, non-regular files, and sensitive mounts. Container deployments expose an explicit read-only file-tail root rather than the entire host filesystem.
Host heartbeats
The host agent can post bounded JSON snapshots to POST /v1/heartbeats. Heartbeats include host state such as load, memory, disks, networking, processes, and container summaries. They power host_state, fleet_state, and correlate_state.
Heartbeat request bodies are capped at 256 KiB. Heartbeat data has short operational retention separate from the main log-retention policy.
AI transcripts
Cortex indexes local and forwarded transcript data from:
Claude Code projects under ~/.claude/projects
Codex sessions and worktrees under ~/.codex/sessions and ~/.codex/worktrees
Gemini CLI chat data under ~/.gemini/tmp
Antigravity desktop's redacted transcript projections under ~/.gemini/antigravity/brain/<session>/.system_generated/logs/transcript.jsonl
Antigravity CLI's redacted transcript projections under ~/.gemini/antigravity-cli/brain/<session>/.system_generated/logs/transcript.jsonl
The Antigravity adapter reads only the narrow redacted JSONL projections above; it does not scan conversation databases or other brain artifacts. The scanner supports incremental checkpoints, parse-error records, bounded chunks, broad-path rejection, and safe recovery from changed files. It extracts normalized transcript rows plus dedicated skill, MCP tool-call, and hook events where those lanes are represented by the provider schema, and reports unsupported lanes honestly.
A satellite agent can send already-parsed records to POST /v1/ai-transcripts, which prevents transcript collection from depending on the database living on the same host as the AI client.
Shell and agent activity
Satellite agents can forward additional operational evidence to the shared server:
POST /v1/agent-commands for deduplicated agent command-spool records
POST /v1/shell-history for parsed Bash, Zsh extended-history, and Atuin records
These records use the same storage and correlation model as the rest of Cortex, which makes an agent change or shell command visible beside the service failure that followed it.
Fleet inventory
Inventory collection builds a redacted fleet snapshot from local files, SSH probes, Docker endpoints, and optional service APIs, then projects safe relationships into the investigation graph.
Investigation and intelligence
Cortex exposes bounded workflows rather than a raw SQL console.
Search and context
FTS5 full-text search with host, app, severity, source, project, session, and time filters
Structured filter-only retrieval for indexed fields
Recent tails and single-row retrieval with raw-frame evidence
Surrounding context around a log ID or timestamp
Host, app, and source-IP inventories
Clock-skew measurement using event and receive timestamps
Time and volume analysis
Bucketed timelines
Ingest-rate and queue-pressure state
Near-duplicate message pattern clustering
Recent-versus-baseline anomaly detection
Side-by-side time-range comparison
Silent-host and silent-stream detection
Database, storage, and runtime statistics
Correlation
Cross-host correlation around a timestamp
AI-session anchor correlation against infrastructure logs
Topic resolution through the entity graph before timeline construction
Host-state correlation across logs, heartbeats, and inventory
Historical incident similarity using FTS5
Deterministic incident context bundles with bounded evidence
Recurring errors
An optional background scanner groups repeating error signatures into durable records. Operators can inspect unaddressed signatures, acknowledge them, revoke acknowledgements, and correlate a signature with logs and graph evidence.
Error detection is disabled by default. When enabled, it scans bounded batches, records lower-severity recurrences without paging, and can notify only above a configured severity floor.
Fleet inventory and graph
Inventory collectors
The native inventory subsystem can collect and normalize evidence from:
Local and remote Compose, reverse-proxy, and AdGuard Home configuration
Local process, storage, project, and raw configuration inventories
SSH sessions to remote fleet hosts
Local and remote Docker endpoints
Tailscale
UniFi
Unraid
Media-stack services and related APIs
SSH collection uses strict host-key verification, bounded concurrency, timeouts, and retry backoff. Sensitive fields are redacted before persistence.
The cache lives under ~/.cortex/inventory by default and includes:
normalized/homelab.json: the typed normalized fleet snapshot
collection-state.json: collector health, timing, and warning state
The map action reads the normalized cache. It does not trigger a collection run and does not return raw config bodies or credential-bearing URLs.
Derived investigation graph
The graph connects canonical entities such as:
Hosts and source identities
Logical services and concrete service instances
Applications and containers
Domains, routes, and endpoints
AI projects and sessions
Error signatures and operational findings
Relationships carry confidence, trust, reason codes, timestamps, and bounded evidence references. The graph supports entity resolution, neighborhoods, topology questions, evidence lookup, and explanation paths.
The graph is a rebuildable projection. Raw logs, heartbeats, inventory records, error signatures, and AI session data remain authoritative. Projection rebuilds use staging tables and a short serialized swap, record watermarks and metrics, and preserve explicit degraded state when refresh fails.
AI session intelligence
Cortex treats AI transcripts as operational evidence, not merely chat archives.
Deterministic session analysis
The shared service layer can:
List and search sessions by project
Measure activity in five-hour usage blocks
Summarize project context
List observed tools and projects
Detect frustration or abuse signals
Group those signals into scored incidents
Correlate AI activity with non-AI infrastructure logs
Extract skill invocations
Extract MCP server and tool-call events
Extract hook configuration and runtime events
Build skill-first, MCP-first, and hook-first investigation bundles
The deterministic query and incident workflows are available through CLI, REST, and MCP.
Guarded local assessments
LLM-backed assessments are deliberately local-only. They run through cortex assess and are not exposed as MCP actions or REST routes because they spawn a local provider subprocess.
CORTEX_LLM selects the provider and optional model for every LLM operation: skill, MCP, hook, and abuse assessments, including dry-run audit metadata. Set CORTEX_LLM=codex (the default, using Codex app-server) or CORTEX_LLM=gemini (Gemini CLI). To pin the model globally, use CORTEX_LLM=codex/gpt-5.5 or CORTEX_LLM=gemini/gemini-3.1-flash-lite-preview. An explicit model in CORTEX_LLM wins over --model; otherwise that existing flag can select a model for one invocation. Empty or unsupported selectors fail without provider fallback.
The selector also works in managed ~/.cortex/.env; process environment values take precedence. Settings are retained after config loading.
The old CORTEX_CODEX_MODEL and CORTEX_HEADLESS_GEMINI_MODEL variables are ignored. Command and authentication-home overrides remain installation settings, not provider selectors. --no-llm needs no provider configuration or credentials.
The shared LLM runner enforces:
A global kill switch
Global and per-action concurrency limits
Per-minute and per-hour rate limits
Per-action circuit breakers and cooldowns
Invocation timeouts
Prompt and output byte caps
An explicit background-enrichment gate, disabled by default
Durable audit records for successes, failures, timeouts, and policy denials
Default guard values allow one concurrent invocation, three per minute and thirty per hour per action, a 120-second timeout, a 1 MiB prompt cap, and a 256 KiB output cap.
Prompt scrubbing is enabled by default. Skill, MCP, and hook event extraction happens before scrubbed transcript text is persisted, so structured operational signals are retained without requiring raw prompt storage.
Alerts and notifications
Notifications are optional and disabled by default. When enabled, Cortex uses Apprise as the delivery bridge and a durable SQLite outbox for retry, deduplication, and dead-letter handling.
Built-in evaluators cover:
OOM kills
Containers exiting nonzero
fail2ban bans
Authelia MFA failures
Disk-fill and storage guardrail pressure
Ingest queue pressure
Complete ingest silence
Heartbeat silence
Silence from previously active continuous streams
The notification subsystem includes:
Configurable evaluator cadence
Per-rule toggles and thresholds
Deduplication windows
Outage-scoped silence keys
Bounded retry with dead-letter state
Recent-firing history
Test notifications
A scheduled daily digest
By default, continuous stream-silence tracking covers UDP/TCP syslog, agent Docker, Docker stream and event records, and managed file tails. Sporadic sources such as transcripts and shell history are intentionally excluded.
Interfaces
CLI
Run cortex --help and command-specific --help for the generated command tree.
Group
Purpose
search, filter, tail
Log retrieval
hosts, apps, entity, graph
Discovery and topology
analysis, correlate, state, stats, timeline
Investigation and analytics
sessions, assess
AI-session queries and local guarded assessments
alerts
Error signatures, acknowledgements, and notification history
ingest, heartbeat
Collectors, agents, file tails, inventory, and heartbeats
The CLI supports direct/local operation and HTTP operation. REST-backed mode is the normal remote path and uses CORTEX_URL plus CORTEX_API_TOKEN.
MCP
Cortex exposes one MCP tool named cortex. Its required action field selects an action from a single authoritative Rust registry. The mechanically generated current count is published in the live coverage inventory.
The runtime schema contains per-action flags, defaults, examples, relative cost metadata, and validation. See docs/mcp/SCHEMA.md for the parameter reference.
Schema for structured incident-style prompt output
ui://cortex/query-widget
Self-contained MCP Apps search widget
The query widget is progressive enhancement. UI-capable MCP hosts can render it; ordinary MCP clients continue receiving the normal text and structured JSON result.
REST
Authenticated JSON routes live under /api/* on the shared HTTP listener. They cover the same major query domains as MCP and add operator workflows for session checkpoints, parse errors, database integrity jobs, backup, checkpoint, and vacuum.
The versioned investigation API lives under /api/v1/* and provides Ask Cortex plus graph entity, neighborhood, explanation, and evidence endpoints for the bundled browser workspace.
REST requires CORTEX_API_TOKEN. Privileged maintenance and file-tail workflows can also require CORTEX_API_ADMIN_TOKEN.
See docs/api.md for the route and response reference.
Browser investigation workspace
The daemon serves a bundled workspace at /app and /app/investigate. It includes:
Runtime and schema status
Fleet and ingest summaries
Ask Cortex investigation requests
An interactive Cytoscape graph canvas
Evidence inspection
A recent-log timeline
The app is embedded into the Rust binary, has no external runtime dependency, uses a restrictive Content Security Policy, and keeps the entered REST bearer token in memory only. The current UI identifies itself as an investigation preview rather than a full general-purpose dashboard.
Claude Code plugin
.claude-plugin/plugin.json packages Cortex as a Claude Code plugin. It declares exactly three surfaces plus configuration:
Key
Points at
mcpServers
plugins/cortex/mcp.json — HTTP transport to ${user_config.server_url}/mcp with a Bearer ${user_config.api_token} header
skills
plugins/cortex/skills/ — twelve skills
userConfig
Server-vs-client mode, server_url, api_token, auth mode and OAuth fields, syslog/MCP bind host and port, retention and storage budgets, Docker ingest, and fleet hosts
The twelve skills are cortex, frustration-assessment, hook-friction-assessment, incidents, logs, mcp-friction-assessment, report, searching-sessions, skill-improvement-assessment, topology, troubleshoot, and version-check.
The plugin registers no Claude Code lifecycle hooks — there is no hooks key and no hooks.json. Setup is explicit: run cortex setup pluginhook (or the plugins/cortex/scripts/plugin-setup.sh adapter) after installing, upgrading, or reconfiguring. just validate-plugin and scripts/validate-marketplace.sh assert the hooks key stays absent, and cargo xtask check-version-sync asserts the manifest carries no top-level version.
GET /health: minimal unauthenticated liveness response for containers and proxies
GET /health/full: detailed health and ingest observability; mounted auth requires CORTEX_TOKEN (OAuth alone is insufficient). Explicit loopback and trusted-gateway policies bypass this check.
Set CORTEX_HEARTBEAT_TARGET and an ingest credential first.
CORTEX_API_TOKEN is query-only and cannot authenticate agent ingest. See the
authoritative macOS heartbeat-agent operator
contract for capabilities, paths,
login semantics, migration, recovery, removal, and delivery proof.
Cortex loads configuration in this order, with later layers winning:
Built-in defaults
A partial config.toml in the current working directory
The managed $CORTEX_HOME/.env file, normally ~/.cortex/.env
Process environment variables
Important defaults
Setting
Default
Syslog bind
0.0.0.0:1514 UDP and TCP
HTTP bind
127.0.0.1:3100
Database path
/data/cortex.db unless setup or environment overrides it
SQLite pool size
8 connections
SQLite page cache budget
128 MiB total
SQLite mmap target
256 MiB
Heavy-read concurrency
1
WAL checkpoint threshold
256 MiB
Log retention
90 days
Logical database limit
1,024 MiB
Recovery threshold
900 MiB, auto-adjusted when the limit is raised substantially
Cleanup cadence
60 seconds
Notifications
Disabled
Recurring-error scanner
Disabled
Central Docker pull
Disabled
Prompt scrubbing
Enabled
Background LLM enrichment
Disabled
Useful environment variables include:
Variable
Purpose
CORTEX_HOME
Managed configuration and operational home
CORTEX_DB_PATH
SQLite database path
CORTEX_RECEIVER_HOST, CORTEX_RECEIVER_PORT
Syslog listener
CORTEX_ALLOWED_SOURCE_CIDRS
Optional syslog sender allowlist
CORTEX_HOST, CORTEX_PORT
Shared HTTP listener
CORTEX_TOKEN
Static MCP, OTLP, heartbeat, and forwarding bearer token
CORTEX_API_TOKEN
REST bearer token
CORTEX_API_ADMIN_TOKEN
Additional REST admin token where required
CORTEX_AUTH_MODE
Static-token or OAuth authentication mode
CORTEX_PUBLIC_URL
Public URL used for OAuth and exposure validation
CORTEX_ALLOWED_HOSTS, CORTEX_ALLOWED_ORIGINS
Host-header and CORS policy
CORTEX_RETENTION_DAYS
Log retention
CORTEX_MAX_DB_SIZE_MB
Logical database storage budget
CORTEX_MIN_FREE_DISK_MB
Optional external free-space write guard
CORTEX_FILE_TAIL_ALLOWED_ROOTS
Managed file-tail root allowlist
CORTEX_DOCKER_INGEST_ENABLED
Enable central Docker pull compatibility mode
CORTEX_SCRUB_PROMPTS
Best-effort AI prompt credential scrubbing
CORTEX_NOTIFICATIONS_ENABLED
Enable notification services
CORTEX_LLM
Shared provider/model selector: codex[/MODEL] or gemini[/MODEL]; default codex
CORTEX_LLM_ENABLED
Global local-assessment kill switch
RUST_LOG
Tracing filter
See docs/CONFIG.md for the complete reference and validation rules.
Authentication and trust boundaries
Cortex intentionally separates transport credentials and capabilities.
Static-token callers receive cortex:read by default.
Set CORTEX_STATIC_TOKEN_ADMIN=true only when the token should also receive cortex:admin.
OAuth mode supports explicit read and admin scopes through the shared authentication layer.
Non-loopback unauthenticated binds are rejected unless an explicit trusted-gateway mode is configured.
REST
CORTEX_API_TOKEN is separate from CORTEX_TOKEN.
REST refuses to mount without an API token.
Sensitive maintenance and file-tail operations can require CORTEX_API_ADMIN_TOKEN in addition to the normal REST bearer token.
Ingest endpoints
Syslog itself is unauthenticated. Use CIDR and network controls.
OTLP logs, metrics, and traces, plus heartbeats, AI transcripts, shell history, and agent-command forwarding, use the Cortex bearer-token policy when configured.
Unauthenticated forwarding endpoints are loopback-only unless the operator explicitly establishes another trusted boundary.
Data handling
AI prompt scrubbing is enabled by default.
Inventory is redacted before it reaches the normalized cache.
MCP and REST return bounded evidence instead of raw credential-bearing config artifacts.
File tails are constrained to configured roots and reject path escapes.
Docker endpoints and SSH keys are treated as privileged infrastructure access.
Container examples mount a dedicated least-privilege SSH directory rather than a user's complete ~/.ssh.
FTS5 external-content indexing with logs_fts for the complete log corpus and a transcript-only ai_logs_fts projection that indexes message text plus provider scope for AI session search
Covering and composite indexes for common filters and timelines
Transactional batch writes
Durable source checkpoints and parse errors
Maintenance job tracking
Online backup support
Integrity checks, checkpoints, and vacuum workflows
The current schema history contains 63 sequential migrations. CI derives this denominator from KNOWN_SCHEMA_VERSION and the migration registry. Migration 63 adds indexed device attribution and a bounded, resumable background history pass that preserves raw authentication and session identity. Migration 60 introduced the transcript-only ai_logs_fts projection. Migration 61 rebuilds that derived index with provider scope and adds an AI-only (timestamp, ai_tool) index so time-bounded session searches can derive a safe rowid floor before FTS walks common terms. On large transcript histories these one-time derived-index rebuilds can hold the startup write transaction while they populate. Subsequent transcript inserts and deletes maintain ai_logs_fts incrementally, while global logs_fts remains the full-corpus search index. Server forwarding receipts have a seven-day replay horizon and are removed when their canonical evidence is deleted. The sender spool is intentionally shorter and bounded: an individual source retains at most 1,024 records or 1 MiB and evicts records older than one day; the aggregate spool retains at most 4,096 records or 4 MiB. Eviction removes the original payload and retains a pending gap marker so evidence loss remains visible. A retry after the server receipt horizon is a new ingestion attempt.
Authoritative and derived data
Authoritative records include:
Normalized logs and raw frames
Hosts and source identities
Transcript source, checkpoint, import, and parse-error records
Heartbeats and component snapshots
Inventory snapshots and collector state
Error signatures and acknowledgement events
Skill, MCP, and hook events
Notification outbox and firing history
LLM invocation audit records
Derived accelerators include:
AI session rollups
Hourly timeline rollups
Inventory statistics
Stream-last-seen state
Entity graph tables and evidence relationships
Derived data can be refreshed or rebuilt from authoritative evidence.
Storage guardrails
The daemon enforces retention and a logical database budget in chunks. It can enter a write-block state before uncontrolled growth damages the host, then resume after recovery thresholds are met. An optional minimum-free-disk guard protects against pressure caused by data outside the Cortex database without deleting Cortex data merely because another application filled the filesystem.
Operational commands:
cortex db status
cortex db integrity
cortex db checkpoint
cortex db backup --help
cortex db vacuum --help
Maintenance operations, including synchronous and background integrity checks, share one process-wide single-flight gate and are separately limited from heavy read queries. Concurrent attempts return a retryable busy response.
Deployment and distribution
Current identities
Artifact
Current identity
Canonical source repository
dinglebear-ai/cortex
Native binary and CLI
cortex
npm launcher
@dinglebear/cortex
MCP Registry server name
ai.dinglebear/cortex
Published OCI image
ghcr.io/dinglebear-ai/cortex:v<version>
The source repository and its published artifacts both live under the dinglebear-ai organization. The legacy jmagar namespace is retired: scripts/check-public-identity.sh fails the build on any tracked file that reintroduces it, and the release workflow derives the MCP Registry OCI identifier from the same REGISTRY/IMAGE_NAME it pushes to, so the two cannot drift apart again.
Container images published before the move remain readable under the legacy namespace for older pinned deployments, but nothing new is pushed there.
Native and npm
The repository ships Linux x86_64, Linux aarch64, macOS arm64, and Windows x86_64 release assets.
The npm launcher supports linux/x64, linux/arm64, darwin/arm64, and win32/x64 and verifies release checksums.
Source builds use Rust edition 2024 and the current stable toolchain in CI.
Docker Compose
The repository includes:
docker-compose.yml for a local build from the checkout
docker-compose.prod.yml for the published image
config/Dockerfile for a non-root Debian runtime image
Manual Compose deployment expects an external Docker network named cortex unless DOCKER_NETWORK overrides it:
The preflight resolves the /backups bind with Docker Compose's own parser and
creates the default or CORTEX_BACKUP_DIR override at mode 0700. Compose is
configured not to create this host path implicitly, preventing root-owned or
overly permissive backup directories.
The Compose files:
Publish syslog on UDP/TCP 1514
Publish HTTP to host loopback by default
Persist the database in a stable named volume or configured bind mount
Run as a non-root UID/GID
Mount inventory SSH credentials and workspace data read-only
Mount a dedicated file-tail root read-only
Use a 2 GiB default memory limit, configurable per host
Include a health check and bounded container logs
For managed installation and repair:
cortex setup check
cortex setup repair
Host agents
Cortex includes setup and runtime support for satellite collection, including:
Heartbeat agents
Docker stream agents
AI-session watch services and timers
Shell-history forwarding
Agent-command forwarding
Completion and debug wrappers
Use cortex setup --help, cortex ingest --help, and cortex heartbeat --help for the exact platform-specific command tree.
Operations
Common operator commands:
# Health and diagnostics
cortex status
cortex doctor
cortex doctor binary
# Managed setup
cortex setup check
cortex setup repair
# Database operations
cortex db status
cortex db integrity
cortex db checkpoint
cortex db backup --help
cortex db vacuum --help
# Compose lifecycle
cortex compose status
cortex compose doctor
cortex compose pull
cortex compose up
cortex compose restart
cortex compose logs
# Collection and inventory
cortex ingest inventory status --json
cortex ingest inventory refresh --json
cortex ingest filetail list
cortex sessions doctor
# Updates and configuration
cortex update --help
cortex config list
cortex completions zsh
The daemon exposes minimal and full health responses, records database maintenance jobs, reports projection and collector degradation, and surfaces ingest queue and write-block state through CLI, REST, and MCP.
Development and verification
Tooling
The repository uses:
Rust edition 2024
mise for pinned development tools
just for common workflows
cargo-nextest for the hermetic Rust suite
cargo-llvm-cov for coverage
cargo-deny for dependency policy
Lefthook and cargo xtask for local release and pre-push checks
A custom soldr-backed Rust compiler wrapper for fast local and CI builds
Common commands
mise install
just dev
just build
just check
just lint
just fmt
just test
just test-doc
just coverage
just coverage-html
just test-live
just validate-plugin
cargo xtask pre-push
just test-live (also just live-smoke) is the canonical fail-closed pull-request subset. It exercises real HTTP JSON-RPC, UDP and TCP syslog ingest, CLI/REST behavior, browser routes, and managed file-tail behavior in a run-owned topology. Run just live-mcp for every registered MCP action; the scheduled aggregate combines all authoritative owner profiles. Specialist profiles are documented in the live qualification guide. Docker collection has separate agent-deployment tests and a mocked Docker HTTP fixture for central pull.
The code-owned registries and runtime schemas are authoritative for command names, actions, routes, scopes, defaults, and validation. Human documentation explains how to operate those surfaces.
Dated design plans and session logs preserve engineering history, not current public-interface authority. Runbooks are operational guidance unless explicitly marked historical; verify them against the current source and configuration.
Current boundaries
Cortex is intentionally opinionated:
It is a single-node SQLite service, not a distributed ingestion cluster.
OTLP logs, metrics, and traces are supported over HTTP/protobuf. OTLP/gRPC is outside the current implementation.
Syslog does not authenticate senders. Network and CIDR controls matter.
The graph is derived evidence, not authoritative configuration state.
MCP and REST expose bounded operations, not arbitrary SQL or unaudited log mutation.
LLM-backed assessments are local-only and require an operator-controlled installation of the provider selected by CORTEX_LLM.
Central Docker pull requires privileged read access to Docker endpoints and is disabled by default.
Inventory quality depends on collector access, SSH trust, optional API credentials, and cache freshness.
The bundled browser app is an investigation workspace preview, not a full monitoring dashboard.
Cortex is built for a homelab or small trusted fleet. Internet-facing or multi-tenant deployment requires additional isolation and policy outside the binary.
License
Original Dinglebear-authored portions of this project are licensed under AGPL-3.0-only. Separate commercial licensing is available for organizations that need terms outside the AGPL. Third-party material remains under its original license. See LICENSING.md.
अभी टूल सूची नहीं है। रिमोट सर्वर को पहले हेल्थ चेक के बाद सूची मिलती है।
FAQ
सवाल
Cortex RMCP कैसे इंस्टॉल करें?
इस पेज पर Cursor, Claude Desktop, Claude Code या VS Code के लिए stdio कॉन्फ़िग कॉपी करें। यह पब्लिशर का इंस्टॉल इस्तेमाल करता है, PlugMCP प्रॉक्सी नहीं।
क्या Cortex RMCP के लिए API key चाहिए?
पब्लिशर ने ये एनवायरनमेंट वेरिएबल बताए हैं: CORTEX_HOME, CORTEX_TOKEN, CORTEX_API_TOKEN, CORTEX_RMCP_VERSION, CORTEX_RMCP_REPO, CORTEX_RMCP_RELEASE_BASE_URL, RUST_LOG।
Cortex RMCP कौन सा ट्रांसपोर्ट इस्तेमाल करता है?
Cortex RMCP stdio पर लोकल चलता है।
Cortex RMCP की पिछली जाँच कब हुई थी?
पिछली जाँच 2026-11-07 को हुई। Rust MCP server for homelab logs, syslog, Docker logs, FTS search, and AI transcript correlation.
