secureclaw and safeclaw

SecureClaw provides OWASP-aligned security policies and skill validation for OpenClaw agents, while SafeClaw complements it by offering runtime monitoring and interception of those agent actions—making them complementary layers of a defense-in-depth security architecture for the same platform.

secureclaw
44
Emerging
safeclaw
39
Emerging
Maintenance 13/25
Adoption 10/25
Maturity 3/25
Community 18/25
Maintenance 10/25
Adoption 5/25
Maturity 9/25
Community 15/25
Stars: 251
Forks: 36
Downloads:
Commits (30d): 0
Language: TypeScript
License:
Stars: 10
Forks: 5
Downloads:
Commits (30d): 0
Language: TypeScript
License: MIT
No License No Package No Dependents
No Package No Dependents

About secureclaw

adversa-ai/secureclaw

SecureClaw - Security Plugin and Skill for OpenClaw OWASP-Aligned

Implements a dual-layer defense combining 56 static audit checks and automated hardening modules with a lightweight 15-rule LLM skill (~1,230 tokens) that runs behavioral detection via external bash scripts rather than consuming context window. Maps comprehensive coverage across OWASP ASI, MITRE ATLAS, and CSA frameworks, detecting threats including prompt injection, credential exfiltration, and supply chain attacks specific to agentic AI architectures.

About safeclaw

DinoMorphica/safeclaw

Security Dashboard for OpenClaw AI Agents - intercept, monitor, and control what OpenClaw does on your system.

Implements a threat detection engine with 200+ patterns across 10 categories and secret scanning, while blocking dangerous commands via pattern-based rules before execution. Built as a monorepo with a Fastify backend using SQLite/Drizzle ORM, Socket.IO for real-time updates, and a React 18 frontend, designed to auto-connect to running OpenClaw agent instances.

Related comparisons

Scores updated daily from GitHub, PyPI, and npm data. How scores work