skills.rest
dsgalkar
dsgalkarCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

vercel-plugin-eval

Runs live eval sessions against the vercel-plugin to verify hook firing, skill injection, and dedup correctness.

Updated Sep 5, 2026
Search Tags:#claude code#wezterm#coverage report#skill injection#hook testing#plugin eval#dedup verification
One-click install
npx skills add https://github.com/dsgalkar/dnyaneshwar_portfolio --skill vercel-plugin-eval-dsgalkar
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: vercel-plugin-eval
Source: https://github.com/dsgalkar/dnyaneshwar_portfolio/tree/main/.agents/plugins/vercel/.claude/skills/vercel-plugin-eval
Command: npx skills add https://github.com/dsgalkar/dnyaneshwar_portfolio --skill vercel-plugin-eval-dsgalkar

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Verifying that a Claude Code plugin's hooks actually fire, inject the right skills, and deduplicate correctly requires launching real interactive sessions, which is tedious and error-prone to do manually. This Skill standardizes that eval workflow with exact commands, monitoring queries, and a structured coverage report. ## Core Features & Use Cases - Live Session Launching: Spawns real Claude Code sessions via WezTerm with the plugin installed through npx add-plugin, avoiding non-interactive modes where hooks never fire. - Hook & Dedup Monitoring: Provides grep-based checks for all 8 registered hooks (SessionStart, PreToolUse, UserPromptSubmit, PostToolUse, SessionEnd) and compares skill injections against atomic claim files to validate dedup. - Coverage Reporting: Defines scenario types (AI chat app, monorepo, edge auth, marketplace) mapped to the 44 plugin skills, plus a .notes/COVERAGE.md report format tracking hook coverage, injection tables, and issues found. - Use Case: After modifying the vercel-plugin's hook logic, run a matrix of scenario prompts across fresh sessions and confirm every expected skill injected exactly once with no missed hooks. ## Quick Start Ask the agent to run a plugin eval session for an AI chat app scenario, monitor the debug log for hook firing and dedup claims, and write the coverage report to .notes/COVERAGE.md.

Frequently Asked Questions about vercel-plugin-eval

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

FAQPage Schema
How do I test that Claude Code plugin hooks actually fire?▼

Launch a real interactive session via WezTerm with the plugin installed using npx add-plugin, then grep the debug log in ~/.claude/debug for SessionStart, PreToolUse, UserPromptSubmit, PostToolUse, and SessionEnd success entries. Hooks do not fire with claude --print or -p.

How do I verify skill injection dedup in a Claude Code plugin?▼

Count skillInjection entries in the debug log and divide by three, since each injection logs three times. Compare that number against the claim files in the session's seen-skills directory under the OS temp dir, which are created atomically with O_EXCL, one per skill.

Why don't hooks fire when testing a Claude Code plugin?▼

Hooks do not fire in non-interactive modes such as claude --print or -p, and sessions must not use --dangerously-skip-permissions. Launch an interactive session through WezTerm with /bin/zsh -ic and install the plugin via npx add-plugin rather than wiring settings manually.

How do I trigger hard-to-reach plugin skills during eval?▼

Some skills require explicit technology references in the prompt because agents do not naturally reach for them. For example, say "use the AI Elements component registry" for ai-elements, "generate components with v0" for v0-dev, or "install the geist font package" for geist.

What should a plugin eval coverage report include?▼

Write results to .notes/COVERAGE.md with a session index of slugs and session IDs, a hook coverage matrix, a skill injection table across the 44 skills, dedup stats comparing injections to claims, and a list of issues found such as bugs or validation gaps.

Related Skills

Explore More Software Engineering Skills→
sharad07072007sharad07072007

benchmark-testing

Creates isolated test projects and launches WezTerm panes running Claude Code to benchmark vercel-plugin skill injection.

Community
Intermediate
mmdmcymmdmcy

tdd

Writes a failing regression test before fixing a bug and validates the fix.

Community
Intermediate
marcoguillermazmarcoguillermaz
3

external-review

Run fresh-context pre-release reviews of CDK with cross-LLM evaluations.

Community
Advanced
cipherstashcipherstash
17

testing-anti-patterns

Detect testing anti-patterns that verify mocks instead of real behavior.

Community
Basic