AI KNOWLEDGE HUB • SUPER EARLY BUILD • OPEN SOURCE SKILLS • CONTRIBUTE VIA PR •

Security Plugins / Runtime Safety

Runtime Safety Plugin

Package untrusted-content handling, dependency audit, secret hygiene, and environment-risk assessment into one conservative security review workflow plugin.

security, runtime-safety, secrets, supply-chain

Install this plugin

Install (Codex)

./bin/skills-hub install --module plugins --entry security/[email protected] --runtime codex

Install (Claude)

./bin/skills-hub install --module plugins --entry security/[email protected] --runtime claude

Install (Generic)

./bin/skills-hub install --module plugins --entry security/[email protected] --runtime generic --target ./my-agent/plugins

Plugin installs also resolve bundled skills, agents, and tools into their native runtime directories. Packaged hooks stay inside the plugin directory.

Codex and Claude installs also generate a runtime-specific plugin manifest inside the installed plugin directory.

This plugin is still experimental. Review bundled components, required secrets, and approval rules before enabling it in a live runtime.

  • Approval rules: human-review-for-live-remediation

Status

Readiness: Experimental

Security reviewed: no

Lifecycle: Active

Install Summary

Installed skills: 4

Installed agents: 0

Installed tools: 0

Packaged hooks: 1

Installed Agents

No bundled agents declared.

Installed Tools

No bundled tools declared.

Metadata

ID: security/runtime-safety-plugin

Latest: 0.1.0

Runtimes: codex, claude, generic

Requirements

Secrets: 0

Approvals: 1

Hooks: 1

Approval rules: human-review-for-live-remediation

Hook names: quarantine-suspicious-instructions