sapper-ai
v0.8.1
Published
AI security guardrails - single install, sensible defaults
Maintainers
Readme
sapper-ai
Single-install AI security guardrails for tool-calling systems.
Quick start
pnpm add sapper-ai
npx sapper-ai initimport { createGuard } from 'sapper-ai'
const guard = createGuard()
const decision = await guard.check({ toolName: 'shell', arguments: { cmd: 'ls' } })Presets
| Preset | Description | |---|---| | monitor | Monitor only - logs threats but never blocks | | standard | Balanced protection with sensible defaults | | strict | Strict enforcement with lower thresholds | | paranoid | Maximum security - aggressive blocking, fail closed, LLM analysis | | ci | CI/CD pipeline - deterministic, fail closed, no LLM | | development | Development mode - permissive, monitor only |
More details: https://github.com/sapper-ai/sapperai#readme
