Review code prompts for destructive operations and security risks using the NEURO API.
npx skills add https://github.com/weebcoder101/dreamcode --skill guardian-ai
Please help me install this Agent Skill. Skill: guardian-ai Source: https://github.com/weebcoder101/dreamcode/tree/main/.dreamcode/skills/guardian-ai Command: npx skills add https://github.com/weebcoder101/dreamcode --skill guardian-ai
This Skill acts as a pre-execution safety net, reviewing every user prompt to prevent accidental destructive operations and catch security issues before they occur.
Use the guardian-ai skill to review a prompt with 'guardian-ai --prompt "delete all users table"'
High-intent search queries and answers about installing and using this skill.
To perform a pre-execution security review, use the guardian-ai Skill. It checks code prompts against the NEURO API to prevent destructive operations, catching security risks and data loss before execution.
Yes, you need a NEURO_API_KEY to run safety checks. The Skill uses the NEURO API to conduct in-depth reviews of user prompts for security risks and destructive operations.
A code safety review handles known dangerous filesystem operations by using hard-block checks. This instantly blocks operations like filesystem deletion to prevent accidental data loss.
Yes, you can customize risk levels for manual approvals in a security review workflow. The Skill allows you to configure risk levels to suit your specific security-critical operations.
The best way to integrate safety checks with sensor gates and code-hardener workflows is through the Skill chain. The Skill is specifically designed to integrate seamlessly into existing security-critical workflows.
Yes, common non-risky commands are automatically approved during a security review. The Skill uses safe pattern checks to instantly allow known safe operations without requiring manual review.
shekerkammaAutomate Google Ads API account discovery, GAQL reporting, and campaign management via authenticated requests.
Phife726Automate code-review requests with base and head SHA tracking.
renatoasseOrchestrate multi-agent squads with checkpoints and human approvals in IDEs.
ils15Audit codebases for vulnerabilities, dependencies, container security, SBOM, PII, and compliance.
