An AI-powered interview coach designed to cover the full job search lifecycle, from initial job description analysis and resume optimization to mock interviews and post-offer negotiation. It targets job seekers aiming for a structured, data-driven, and personalized approach to interview preparation, offering adaptive coaching that goes beyond generic advice by diagnosing root causes of performance gaps and building a dynamic storybank.

How It Works

The core approach is an AI-driven system that analyzes user performance across five key dimensions: Substance, Structure, Relevance, Credibility, and Differentiation. It diagnoses specific root causes for weak spots, such as "status anxiety" or "narrative hoarding," and employs a decision tree to direct users to targeted drills. The system uniquely adapts its coaching based on interview format (behavioral, system design, panel) and user patterns, incorporating multi-format transcript analysis and a dynamic storybank with portfolio optimization. Its effectiveness is further refined through outcome calibration, where practice scores are correlated with real interview results.

Quick Start & Requirements

• Primary install/run command: Clone the repository, rename SKILL.md to CLAUDE.md (for Claude Code) or AGENTS.md (for OpenAI Codex), open the folder in the respective environment, and run the kickoff command.
• Non-default prerequisites: Requires a paid Claude plan (for Claude Code) or a paid ChatGPT plan (for OpenAI Codex). Also compatible with Claude Code (terminal) or any environment with file system access.
• Estimated setup time: Coaching begins in under 2 minutes after providing a resume.
• Links: Official GitHub repository: https://github.com/noamseg/interview-coach-skill.git.

Highlighted Details

• Comprehensive 23 commands cover the full job search lifecycle, from JD analysis to post-offer negotiation.
• Adaptive coaching scores answers on five dimensions, diagnoses root causes, and tailors drills to individual bottlenecks.
• Advanced transcript analysis supports multiple formats (Otter, Zoom, etc.) with format-specific scoring and anti-pattern detection.
• Storybank management includes portfolio optimization, rapid retrieval drills, and narrative identity extraction for coherent self-presentation.
• Outcome calibration continuously refines coaching by correlating practice scores with real interview results and detecting scoring drift.
• Optional "Directness Level 5" Challenge Protocol applies rigorous feedback lenses like assumption audits and pre-mortems.

Maintenance & Community

Created by Noam Segal. No explicit community channels (e.g., Discord, Slack) or detailed maintenance roadmap are provided in the README.

Licensing & Compatibility

MIT License. This license is permissive and generally allows for commercial use and integration into closed-source projects.

Limitations & Caveats

The "Directness Level 5" features, including the Challenge Protocol, are only accessible at the highest feedback setting. The system's efficacy is contingent on the quality and completeness of user-provided data, such as resumes, transcripts, and logged interview outcomes. Outcome calibration effectiveness relies on users consistently logging real-world results.