---
title: "smyrick-skills"
description: "Shane Myrick's reusable workflows for research, planning, code review, and writing."
canonical: https://agentpluginsdirectory.com/plugins/smyrick-skills
last-updated: 2026-10-21
---

# smyrick-skills
Shane Myrick's reusable workflows for research, planning, code review, and writing.
- Slug: smyrick-skills
- Publisher: Shane Myrick
- Repository: https://github.com/smyrick/skills
- Manifest: plugin.json
- Version: 1.0.2
- License: MIT
- Category (editorial): productivity
- Skills: 11 (codebase-summary, human-review-pr, human-reviewable-code, humanize-text, manage-product-glossary, mock-interview, personal-research-and-plan, research-and-plan, research-orchestrator, shorten-response, write-a-prd)
- MCP servers: 0
- Stars: 1
- Repository created: 2026-03-19
- Repository last pushed: 2026-10-08
- Publisher type: User
- Listing: https://agentpluginsdirectory.com/plugins/smyrick-skills
- Schema: https://agent-plugins.org/schemas/1.0.0/plugin.schema.json

## What smyrick-skills does, in the publisher's words

A personal library of reusable AI-agent workflows. Each skill has a portable SKILL.md core. OpenAI adapters supply interface metadata; the repository invocation policy also generates a Claude Code collection. The entire library ships as the smyrick-skills plugin, with portable Agent Plugins, Codex, and Claude Code packages.

From the project README, punctuation lightly normalized. Full text: https://raw.githubusercontent.com/smyrick/skills/HEAD/README.md

## Skills

- codebase-summary: Create source-grounded codebase architecture documentation with optional self-contained HTML diagrams.
- human-review-pr: Help Shane review a PR or local diff with a compact brief, consequential code and product decisions, a light architecture review, and guided discussion calibrated to his background.
- human-reviewable-code: Shape coding requests, ongoing implementations, and long AI conversations into coherent, verifiable chunks with fresh-task-ready handoffs and scannable PR documentation.
- humanize-text: Rewrite prose to remove generic AI-writing patterns while preserving facts, meaning, and the writer's voice.
- manage-product-glossary: Create and maintain scoped PRODUCT_TERMS.md glossaries that connect product language, domain rules, and code references.
- mock-interview: Run a source-grounded mock interview with stage-specific practice, evidence-based feedback, and optional session notes.
- personal-research-and-plan: Research a non-code decision and produce a concrete recommendation and action plan. Select direct research, focused read-only leaf scouts, or an internal durable research subworkflow based on the evidence needed.
- research-and-plan: Research a code change and produce a decision-complete implementation plan. Select direct inspection, focused read-only leaf scouts, or an internal durable research subworkflow without modifying code.
- research-orchestrator: Coordinate authorization-gated durable research for another skill or user request. Use when planning or decision work needs multiple focused passes, persistent findings, recovery, or a reusable evidence handoff.
- shorten-response: Apply a concise coworker-style response mode that preserves technical depth, caveats, and required safety communication.
- write-a-prd: Create a product requirements document through bounded discovery, codebase context when available, and explicit product decisions.

Descriptions come from the frontmatter of each SKILL.md, punctuation lightly normalized.