TheArchitectit/agent-guardrails-template

canary-tokens

Insert honeypot tokens into sensitive files to detect if an agent leaks their contents in output

Idioma do texto original: inglês

content-safety-filtering

Topic-based content filtering to block harmful or unauthorized content in agent output

Idioma do texto original: inglês

guardrails-docs-access

Access and read guardrails skill documentation, configuration examples, and usage guides at runtime.

Idioma do texto original: inglês

pi-guardrails-core

Available guardrail tools and automatic enforcement behavior for pi agents

Idioma do texto original: inglês

pi-guardrails-dashboard

How to use and interpret the guardrails panel and status bar in the pi TUI

Idioma do texto original: inglês

prompt-injection-defense

Detect and block prompt injection attacks in tool input (bash commands, write/edit content) before execution

Idioma do texto original: inglês

output-security-secret-scanning

Prevent sensitive data (secrets, credentials, PII) from leaking in agent output

Idioma do texto original: inglês

policy-configuration-hierarchy

How guardrail policies cascade from organization through team to project level

Idioma do texto original: inglês

sandbox-isolation

Run untrusted or risky commands in Docker-based sandbox with resource limits and network isolation

Idioma do texto original: inglês

tool-permission-control

Per-tool permission levels (auto/ask/blocked) that gate which tools the agent can use

Idioma do texto original: inglês

commit-validator

Validates git commits follow COMMIT_WORKFLOW.md standards: AI attribution, single focus, no secrets

Idioma do texto original: inglês

env-separator

Enforces TEST_PRODUCTION_SEPARATION.md: production code first, separate instances, no data mixing

Idioma do texto original: inglês

guardrails-enforcer

Enforces the Four Laws of Agent Safety automatically on all operations. Halts on uncertainty.

Idioma do texto original: inglês