From agent-guardrails
Enforces strict separation between test and production environments. Halts operations and prompts user verification when separation cannot be confirmed, preventing data leaks and credential exposure.
How this skill is triggered — by the user, by Claude, or both
Slash command
/agent-guardrails:env-separatorThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Enforce strict separation between test and production environments.
Enforce strict separation between test and production environments.
Before creating test code or running tests:
If you cannot verify environment separation:
When running in pi, environment separation is supported by the @architectit/pi-guardrails extension:
guardrail_mcp with action check_test_prod_separation for explicit verificationSee [[output-security]] and [[canary-tokens]] for details.
Verify that test and production environments are properly separated for the current operation. Check environment boundaries, credentials, and data flows. Halt if separation cannot be guaranteed.
docs/standards/TEST_PRODUCTION_SEPARATION.md - Full environment rulesskills/production-first/SKILL.md - Production-first mandatedocs/AGENT_GUARDRAILS.md - Core safety protocolsnpx claudepluginhub thearchitectit/agent-guardrails-templateEnforces creating and validating production code before test or infrastructure code. Useful for teams wanting to prevent test-first or infrastructure-first patterns.
Enforces AI agent safety in production: 4-layer input validation, output filtering for secrets/dangerous code, scope boundaries on operations/files/docs, and readiness checklist.
Manages full lifecycle of secrets and environment variables: decides placement (constant, .env, CI secret, env var), scaffolds .env.example/.gitignore, add/update/rotate/remove/migrate/audit/provision across envs. Language-agnostic.