skills.rest
Trigent-Software-Pvt-Ltd
Trigent-Software-Pvt-LtdOFFICIAL
Explore all skills by this author →
GitHub

ai-pair-programmer

Queries Grok, ChatGPT, and Gemini APIs for second opinions on code, plans, and architecture decisions.

Updated Jul 22, 2026
Search Tags:#code review#grok#gemini#pair programming#architecture review#chatgpt#second opinion
One-click install
npx skills add https://github.com/Trigent-Software-Pvt-Ltd/trigen-aidlc --skill ai-pair-programmer-trigent-software-pvt-ltd
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ai-pair-programmer
Source: https://github.com/Trigent-Software-Pvt-Ltd/trigen-aidlc/tree/main/plugins/pair-programming/skills/ai-pair-programmer
Command: npx skills add https://github.com/Trigent-Software-Pvt-Ltd/trigen-aidlc --skill ai-pair-programmer-trigent-software-pvt-ltd

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve? When making substantive design or implementation decisions, a single AI perspective can miss edge cases or better alternatives. This Skill sends your proposed solution, code, or plan to external AI providers (Grok, ChatGPT, Gemini) and returns their critiques so you can synthesize multiple viewpoints before committing. ## Core Features & Use Cases - Multi-Provider Reviews: Query one provider or run Grok, ChatGPT, and Gemini in parallel with --provider all, then compare points of agreement and disagreement. - Flexible Content Sources: Review files, git diffs via stdin, inline content, or architecture proposals without any code files using --proposal. - Typed Reviews: Tailor feedback with review types for plans, code, architecture decisions, diffs, and high-level summaries. - Use Case: Before implementing OAuth2 authentication in a Go API, send your proposed middleware design to Grok and Gemini, evaluate their concerns about token handling, and finalize your approach with higher confidence. ## Quick Start Ask the assistant to review your authentication implementation plan with Grok and Gemini for a second opinion.

Frequently Asked Questions about ai-pair-programmer

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I get a second opinion on my code from multiple AI models?▼

Run the pair_review.py script with --provider all or a comma-separated list like grok,gemini, passing your code via --file or --files. Providers are called in parallel and results are returned with per-provider attribution plus synthesis guidance.

How do I review an architecture decision without code files?▼

Pass your approach with --proposal along with --app-context and --context describing the tech stack and problem. The proposal itself becomes the reviewed content, so no files are required for plan or architecture reviews.

Which API keys are needed for Grok, ChatGPT, and Gemini?▼

Set XAI_API_KEY for Grok, OPENAI_API_KEY for ChatGPT, and GEMINI_API_KEY for Gemini. Keys can also be stored in config.json, but environment variables take priority. Only configured providers are used with --provider all.

Can I override the default model for each AI provider?▼

Yes, use the --model CLI flag, set GROK_MODEL, OPENAI_MODEL, or GEMINI_MODEL environment variables, or edit config.json. Priority order is CLI flag, environment variable, config file, then built-in defaults.

Why does the script fail with 'No content to review'?▼

The script requires one content source: a positional argument, --file, --files, --proposal, or piped stdin. Providing only --context and --app-context without content triggers this error; add --proposal for reviews without code files.

When should I skip getting an AI second opinion?▼

Skip it for trivial syntax fixes, one-liners, pure styling tweaks, or when you are fully confident, to avoid unnecessary API costs. Reserve it for substantive decisions like plans, refactors, and architecture trade-offs.

Related Skills

Explore More Software Engineering Skills→
OutlineDrivenOutlineDriven
5

resolve

Validate code review comments and propose multiple architectural solutions.

Official
Advanced
YazanKittanehYazanKittaneh

review

Enforces a standardized PR review checklist covering tests, linting, documentation, and descriptions before merging or opening pull requests.

Community
Intermediate
vtgiang-dotcomvtgiang-dotcom
3

requesting-code-review

Verify staged Git changes with security scans and baseline checks before commits.

Community
Intermediate
FlorianBruniauxFlorianBruniaux
5.9k

ci-status

Displays current CI pipeline status for the active branch on GitLab or GitHub.

Community
Basic