Route frontend and backend code reviews to Gemini or Codex via superspec/project.yaml.
npx skills add https://github.com/HankLiu447/SuperSpec --skill external-review
Please help me install this Agent Skill. Skill: external-review Source: https://github.com/HankLiu447/SuperSpec/tree/main/skills/external-review Command: npx skills add https://github.com/HankLiu447/SuperSpec --skill external-review
This skill enables automated external AI code reviews of frontend and backend work, helping teams identify design flaws, security gaps, and implementation issues before shipping.
Trigger an external AI review after completing implementation. Ensure superspec/project.yaml contains the review settings, then determine the task type (frontend or backend), run the appropriate provider script, perform hallucination checks, apply validated fixes, and re-submit if necessary.
High-intent search queries and answers about installing and using this skill.
To run external AI code reviews, complete implementation first, configure superspec/project.yaml with provider settings, identify the task type, execute the appropriate provider script, and apply validated fixes. This workflow ensures traceable review results for both frontend and backend work.
The hallucination check validates and documents AI suggestions before applying any changes. This structured process ensures safety and relevance by verifying that Gemini and Codex outputs are accurate and applicable to your specific frontend or backend implementation context.
No, external review routing assigns Gemini specifically for frontend tasks and Codex for backend tasks. This provider-specific configuration is guided by superspec/project.yaml to ensure appropriate AI review capabilities match the technical domain of your work.
Configure superspec/project.yaml with provider-specific settings for each task type before triggering an external review. This file guides the routing logic to select Gemini for frontend tasks and Codex for backend tasks, ensuring structured and traceable review delivery.
Yes, re-submission is part of the workflow if necessary. After performing hallucination checks on AI suggestions and applying validated fixes from the external review, you re-submit the code to ensure all identified design flaws, security gaps, and implementation issues are resolved.
Trigger an external AI code review after completing implementation of your frontend or backend work. This timing allows the review to identify design flaws, security gaps, and implementation issues before shipping, providing a second opinion when your team needs validation.
find-xposed-magiskConfigure server-side environment variables for default user settings across domains.
cahyaongAudit codebases for quality, security, performance, and architectural issues.
majedsiefalnasrDiagnose pre-commit and pre-push failures across PHP and JS/TS toolchains.
angusjfwReview pull requests with a guided, mode-aware process using git and GitHub CLI.
