
caracal
🐾 Authority, not credentials, for AI agents: policy-approved actions, delegation that can only narrow, instant revocation, tamper-evident audit.

🐾 Authority, not credentials, for AI agents: policy-approved actions, delegation that can only narrow, instant revocation, tamper-evident audit.

LLM-driven security-testing agent that runs a continuous tool-calling loop against a target, with 160+ recon, web, cloud, AD and post-exploitation…

Rust attack-path analysis for AD, ADCS and Entra ID

The authorization layer CVE-2026-93328 was missing: scoped grants, path confinement, and signed receipts in front of MCP storage tools. CVE…

Zero-trust reverse proxy and ABAC authorization gateway in Go with embedded OPA policy evaluation and distributed token revocation.

Local static security scanner for Next.js, Supabase and Stripe apps: 24 AST-based checks for leaked keys, missing RLS and grants, unsafe webhooks.…

Cross-platform Electron GUI for the Sliver C2 framework, providing session and beacon dashboards, payload generation, listeners, loot, and cloud…

Enumerate the permissions associated with AWS credential set

Open-source gateway that secures, governs, and observes AI agents' MCP tool calls and LLM traffic, with API-key authentication and an admin console…

Defensive research tool that documents observable API endpoints and user agents of offensive tooling targeting Microsoft Entra ID, supporting…

OneDrive as a covert C2 transport for Cobalt Strike

Sandboxed runtime for autonomous AI agents with declarative YAML policies enforcing filesystem, network, and process constraints, plus endpoint-bound…

Policy-governed LLMSecOps framework providing AST-based SAST, secret scanning, supply-chain and multi-cloud CSPM checks, AI-BoM generation, and CI/CD…

Autonomous red-team engagement platform with MITRE ATT&CK module orchestration, DAG attack-path solving, OPSEC controls, encrypted credential vault,…

The exploit server for out-of-band findings. Point a target at a domain you own. Every HTTP request and every email it sends back lands in a…

The system of action for AI-native cybersecurity—where intent becomes governed execution, evidence becomes operational memory, and every operation…

Enforce least-privilege delegation for AI agents with signed, scoped credentials. Grant sub-agents narrow capabilities and resources, verify actions…

Cloud dead-drop C2 framework — RSA-4096 + AES-256-GCM, 5 cloud providers, Rust-only agents, P2P mesh, persistence engine, credential harvesting