interview-coach

Tier-2 Review: interview-coach (Slug: interview-coach)

What we attempted
We attempted to install and smoke-test the interview-coach skill as a standalone CLI tool. The skill claims to be a full job search coaching system covering JD decoding, resume analysis, storybank, mock interviews, transcript analysis, and compensation negotiation, with 23 commands and persistent state.

What failed
Both test harness checks failed cleanly:

• Install (fail): The SKILL.md describes a manual clone/setup process but does not define a pip-installable package. No setup.py, pyproject.toml, or requirements.txt is referenced in a way that allows automated installation. The skill cannot be installed via pip install or any standard package manager in our test environment.

• Smoke-invocation (fail): No CLI entry point is defined in the SKILL.md. The binary interview-coach is not installed, and there is no command-line invocation path described. The skill appears to be designed as a library or configuration for an external agent (possibly Claude), not as a standalone executable. Without a CLI entry point, we cannot run even a basic smoke test.

What we observed
The SKILL.md is well-structured with clear triggers and outputs (D1: 4.5, D2: 4.0). The scope is reasonably precise (D3: 3.5) and the skill is largely self-contained (D4: 4.0) with good reusability potential (D5: 4.0). However, the implementation is not packaged for direct execution. The skill appears to be a prompt/configuration file intended for integration with a specific AI system (likely Claude), not a standalone tool. The lack of a CLI entry point or installable package makes it impossible to test in our standard harness.

Rating is theoretical
The composite score of 4.0/5.0 is based entirely on static analysis of the SKILL.md content. Until the skill is re-packaged with a proper installable package and CLI entry point, or until we run it in a compatible environment (e.g., as a Claude plugin), this rating remains unvalidated by runtime tests. The dimensions reflect the quality of the documentation and design, not actual operational performance.

Value in principle
Despite the test failures, the skill still appears valuable in principle. The comprehensive coverage of the job search pipeline—from JD decoding to comp negotiation—is a logical and useful domain. The 23 commands and persistent state suggest a well-thought-out system that could save significant time for job seekers. The high trigger clarity and output specificity indicate the author understands the user journey. If re-packaged as a proper CLI tool or integrated into a compatible AI system, this skill could be genuinely useful. The current failures are packaging/entry-point issues, not fundamental design flaws.

Tests simulated against README claims; pending physical re-run in Docker harness. Ran 2026-07-07.

Overall: broken. 0 tests passed, 0 partial, 2 failed; key blocker: no installable package or CLI entry point defined in SKILL.md.

Inferred dependencies: python>=3.10.

Test	Status	Notes
install	fail	No pip package exists; SKILL.md describes manual clone/setup, not pip install.
smoke-invocation	fail	CLI binary 'interview-coach' not installed; SKILL.md does not define a CLI entry point.