Detects prompt injection by its effect on a sacrificial canary model, not just pattern matching: untrusted input hits a powerless model first, a behavioral check reads the residue, and it returns block, flag, or pass before your primary model acts. Inbound preflight sensor, not a guarantee.

DeepSec — AI Security Offense & Defense Platform. Shield audits AI-generated code for hallucinated packages, missing safeguards & AI pattern errors in real time. Spear automates authorized penetration testing with 40+ skill packs, from recon to PoC.

Automatic provider/model fallback chains for DeepSeek Harness agents when LLM requests keep failing (retry exhausted, auth, quota, rate limit)

掌上 dsh — the standalone mobile surface for the dsh web GUI: scan-to-pair device trust, the /m/ phone UI, the /m/api RPC channel with realtime SSE mux bridge, and a polished desktop pairing panel. Extracted from dsh-remote-web-ui (M1, behavior-identical)

Real-time secret-leak guardrails for AI coding agents (Claude Code, Codex), Git hooks, and CI.

Audit a website with the squirrelscan CLI and fix the findings in code. Runs SEO, performance, security, technical, content, accessibility, and 15 other rule categories (260+ rules), returns an LLM-optimized report, then drives an iterative fix loop, mapping issues to source files, applying fixes, a

Create, verify, sign, publish, and run dossiers — the open standard for versioned, signed agent skills. Host them anywhere and verify who wrote them. The dossier ⇄ Claude Code skill bridge (install-skill / skill-export).

Use when writing Playwright tests, fixing flaky tests, debugging failures, implementing Page Object Model, configuring CI/CD, optimizing performance, mocking APIs, handling authentication or OAuth, testing accessibility (axe-core), file uploads/downloads, date/time mocking, WebSockets, geolocation,

CVE hunting harness for Claude Code - 20 skills, 5-agent team, systematic vulnerability research with false positive elimination

Deterministic runtime safety for AI agents: scan PII, gate tool actions, verify LLM output.