Intelligence
Threat Database
Supply chain threat intelligence for AI agent code. Sigil's curated threat database tracks malicious packages, attack campaigns, and detection signatures across npm, PyPI, Hugging Face, RubyGems, and Go.
Threat categories
Rule counts are from the core signature pack v1 shipped with the CLI (cli/packs/core/v1, 266 rules across all phases; Phase 6 uses 13 provenance heuristics).
Install Hook Exploits
Malicious code in setup.py cmdclass, npm postinstall, and Makefile install targets that execute on package install.
Credential Exfiltration
Packages that access environment variables, SSH keys, AWS credentials, or API keys and transmit them to external servers.
Obfuscated Payloads
Base64-encoded execution, hex string decoding, String.fromCharCode chains, and minified backdoors designed to evade manual review.
Network Exfiltration
Outbound HTTP calls, webhook callbacks, socket connections, ngrok tunnels, and DNS tunneling hidden in package code.
Dangerous Code Patterns
Use of eval(), exec(), pickle.loads(), subprocess with shell=True, __import__(), and child_process.exec in unexpected contexts.
Provenance Anomalies
Packages with single-commit histories, no verifiable author, binary blobs, hidden files, or filesystem manipulation.
Tracked campaigns
Data Source: NOMARJ/sigil api/data/known_threats.json (133 entries, 18 campaigns, compiled from Unit42, Datadog Security Labs, CISA, JFrog, Socket.dev, ReversingLabs and others). Sample Size: 5 most recent campaigns shown. Limitations: curated from public research, not from Sigil scan results; last updated 2026-02-20.
CAMP-2025-0022025-11-24Shai-Hulud Wave 2
Second wave using preinstall hooks; 796 packages, 27K+ malicious repos, 14K+ secrets exposed
CAMP-2025-0032025-10-01npm Telegram Bot Stealers
Credential theft via fake Telegram bot packages with Vidar malware
CAMP-2025-0012025-09-15Shai-Hulud Wave 1
Self-propagating npm worm using postinstall hooks; compromised 180+ packages with 2.6B+ weekly downloads
CAMP-2025-0062025-07-01SilentSync RAT
Python RAT with remote command execution, file theft, and screen capture
CAMP-2025-0042025-04-01VVS Discord Stealer
Pyarmor-obfuscated MaaS stealer targeting Discord credentials
How threat intelligence works
Detect
Automated scanners and community reports identify malicious packages across npm, PyPI, and git.
Analyze
Threats are classified by attack vector, severity, and ecosystem. Signatures are added to the detection engine.
Protect
Signed-in users sync the community threat database during scans. Pro and Team add full threat intelligence and AI finding investigation.