Agent Claude Code
SafeClaw tracks OpenClaw AI agent activities using a dual-source approach because the gateway WebSocket does NOT broadcast tool call events.
15 tagged access control, measured the same way as everything else here.
Browse within: approval-workflow 7audit-log 7compliance 7data-governance 7database-security 7
Agent Claude Code
SafeClaw tracks OpenClaw AI agent activities using a dual-source approach because the gateway WebSocket does NOT broadcast tool call events.
Agent Claude Code
You are a specialist agent for publishing the safeclaw npm package. Follow these instructions exactly.
Agent Claude Code
The Skill Scanner provides proactive, pre-execution security analysis of markdown skill definitions. While SafeClaw's runtime Threat Analysis Engine (TC- categories) classifies activities as they happen, the Skill Scanner inspects skill files before\ they're loaded — catching hidden threats that are invisible to human…
Agent Claude Code
Backend/engine code specialist reviewer for an AccessFlow change. Reads the backend/ and engines/ portion of a branch cold and checks it against CLAUDE.md's backend rules and the backend/engine pattern checklists — Modulith boundaries, Java idioms, Flyway, scheduled jobs, the Security Rules, test parity. Returns…
Agent Claude Code
Explains what an AccessFlow change actually does. Reads a branch or PR diff cold and returns a structured briefing — the one-paragraph what, the why, a per-area breakdown of the real changes, the interface surface it moves (endpoints, enums, migrations, config knobs, events), and where a reviewer should look hardest.…
Agent Claude Code
Security specialist reviewer for an AccessFlow change. Reads a branch or PR diff cold through a security lens only — proxy bypass, row-security failing open, self-approval and tenant scoping, credential handling, the auth surface, SSRF, audit tamper-evidence, plugin supply chain — and returns Blockers / Concerns /…
Agent
Runs the direct SubjectBroker parent control without delegation.
Agent
Runs the SubjectBroker VS Code subagent isolation controls.
Agent
Runs the ordinary/default-child unsafe-inheritance control.