Hide a site domain inside a visible name using invisible, passphrase-keyed Unicode. If the name leaks, find out which site leaked it. Single self-contained HTML file, no dependencies.
-
Updated
Jul 10, 2026 - HTML
Hide a site domain inside a visible name using invisible, passphrase-keyed Unicode. If the name leaks, find out which site leaked it. Single self-contained HTML file, no dependencies.
Plant a decoy SSH key and get alerted the moment someone uses it.
CanaryProbe 诱饵探针:给你读不到源码的闭源 coding agent 埋下诱饵主机名和假凭证,一旦二进制去解析或连接它,就在运行期立即告警并留下审计证据 —— 把「它有没有偷偷外联」从一次次反编译,变成一个可核验的 yes/no 事件。
Self-hosted honeytoken system, plant decoy AWS keys, .env secrets, and bait files, then get alerted the moment they're touched or reused.
Client-side canary token generator — web bugs, decoy PDFs, fake credentials & .env bait for breach and intrusion detection. 100% static, GitHub Pages ready.
Output-integrity canary for Claude Code: the model ends every response with a per-user secret token; a Stop hook flags its absence (truncation / injection / drift). Non-blocking by design.
Badger, a stubborn URL honeypot and canary-token tool. Innocent-looking links that quietly note down everyone who opens them, and exactly how.
canary-trap: catch prompt injection in Hermes Agent by the decoy credential it tries to steal. Blocks the call, locks the session, alerts you. Stores no conversation content.
HMAC-signed zero-width invisible canary token generator and output response leak auditor for LLM guardrails
🔔 Multi-source honeypot aggregator: Cowrie SSH + HTTP honeypot + canary tokens → unified JSON API with AbuseIPDB enrichment & real-time alerts
Honeyfiles and canary tokens for AI coding agents. Plants decoy credentials your agents should never touch, then trips on any read or exfiltration.
To associate your repository with the canary-token topic, visit your repo's landing page and select "manage topics."