XORCISM-AI/XORCISM
XORCISM is a self-hosted, open-source enterprise risk/exposure management platform (Node.js/TypeScript over SQLite/PostgreSQL/MySQL) that unifies asset inventory, vulnerability management, threat intel (STIX/TAXII), GRC (ISO 27001, NIST CSF/800-53, EBIOS Risk Manager, NIS2, DORA), incident management, and offensive/defensive testing into one continuously recomputed per-asset and enterprise risk score. It chains recon/exploitation tool connectors (nmap, nuclei, Metasploit, sqlmap) into attack-path/choke-point graphs weighted by fusion exploitability (EPSS+CVSS+KEV+public exploits), validates detection coverage via a purple-team loop against 3,750+ Sigma rules and MITRE ATT&CK/D3FEND, and quantifies ransomware impact in dollar terms via a FAIR-style model — with local-only Ollama AI assistants for triage, hunting, and reporting.
What it is
XORCISM's README, beyond the feature catalog already summarized, carries the concrete deployment constraints and licensing status an engineer needs before adopting it. The server runs on port 9292; first boot prints a one-time admin (admin@xorcism.local + random temp password) to the console.
Preconditions / Affected
- Node 20 only.
better-sqlite3is a native module with prebuilt binaries for Node 20; the page states Node 23+/24 break the ABI. Builds run on any Node, but the runtime needs 20. Windows without a system Node can use the bundledtools/nodejs/node.exe. - SQLite databases must live OUTSIDE OneDrive. The page warns OneDrive "replaces files under open handles and corrupts WAL journals." The code tree can sit under OneDrive; the DBs cannot.
What to do
Fresh setup:
cd xorcism_ts
npm install
npm run build
$env:DB_DIR = "C:\Users\$env:USERNAME\XORCISM_databases" # outside OneDrive
npm start # → http://localhost:9292/loginDocker:
docker compose up -d --buildSQLite persists in the xorcism-data volume (DB_DIR=/data); bind-mount existing DBs instead.
Portability to a server DB engine is opt-in via env var, migrated with a script:
$env:XORCISM_DB_ENGINE = "postgresql" # sqlite (default) | postgresql | mysql | mariadb
python tools/migrate_db.pyHardening/integration env vars worth knowing: XORCISM_ALLOW_REGISTER=0 disables public self-registration; XOR_CVE_IMPORT=0 / XOR_CVE_MATCH=0 disable the hourly NVD importer/matcher; NVD_API_KEY raises the NVD rate limit. API connectors take credentials only via environment variables, never the UI (e.g. QUALYS_API_URL/_USER/_PASSWORD, CALDERA_URL/_API_KEY).
The REST API (/api/v1, OpenAPI at /api/v1/openapi.json) is read-only, tenant-scoped, keyed with Authorization: Bearer xor_… (keys stored SHA-256, carry scopes + optional expiry). HMAC-signed webhooks (X-XORCISM-Signature) fire on incident.created/updated and asset.updated.
Caveats
- Licensing is not actually established. The page calls it "open-source" but says to "see xorcism.ai for licensing terms" and admits no
LICENSEfile exists in the repo — "add a LICENSE file to make the terms explicit." For enterprise adoption this is unresolved; absent a license, default copyright applies. - Not affiliated with or endorsed by MITRE, ANSSI, or OASIS despite integrating their frameworks.
- Provided "as is," no warranty; you are responsible for authorization before running any offensive/scanning connector. Live-scan playbooks are ROE-enforced and capability-gated, but the safe/live distinction is yours to configure.
Organisations named here
- SecureCoders
Technologies named here
- XORCISM
- Node.js
- SQLite
- Nmap
- PostgreSQL / MySQL / MariaDB
- nuclei
- nikto
- whatweb
- wpscan
- sqlmap
- OpenVAS
- Metasploit
- Nessus
- Qualys
- Rapid7
- Caldera
- Dependency-Track
- OSV-Scanner
- depx
- Wiz
- Lacework
- Sysdig
- Aikido
- Burp Suite
- Splunk
- Elastic Security
- Microsoft Sentinel
- QRadar
- SAINT
- OpenCVE
- Microsoft Entra ID
- MISP
- OpenCTI
- SOC Prime
- CrackMapExec
- SSLyze
- subfinder
- theHarvester
- Shodan
- HIBP
- httpx
- WPProbe
- w3af
- YARA
- Rustinel
- Ollama
- NeMo
- LLM Guard
- Llama Guard
- Lakera
- Exploit-DB / SearchSploit
- CIRCL
- OSV
- SOCRadar IOC-Radar
- winpmem/avml
- TypeScript
- MITRE ATT&CK
- Sigma
Techniques named here
- MITRE ATT&CK
- D3FEND
- CAPEC
- A3M
- STRIDE
- Attack path / choke point graph traversal
- Ransomware TTP replay
- Purple-team / adversary emulation
- Kill chain phases
- Attack chaining / playbook auto-follow-on scanning
- Honeypot deception
Threat actors named here
- APT29
- AI-enabled threat actors
Related in GRC

Orca Security: NIST AI Risk Management Framework (AI RMF) Explained

FOSSA: CISA Releases the 2026 SBOM Minimum Elements
Teach Yourself to Phish | Huntress
