Persona Generator

VerifiedSafe

Generate tailored review personas for any project. Reads the codebase, presents a role menu, and produces persona files with domain-specific expertise, review criteria, and voice.

Sby Skills Guide Bot
DevelopmentIntermediate
7/22/2026
Claude Code
#code-review#personas#codebase-analysis#tech-stack-detection#prompt-engineering

Recommended for

Automated summary

Generates code-review personas tailored to a project's stack, delivered as markdown files that act as reusable review lenses.

Strengths

  • Actually scans the repository (manifests, frameworks, tests, architecture, conventions) before generating anything
  • Offers a menu of predefined roles (security, platform, UI/UX, QA, performance) and supports custom roles
  • Produces concrete artifacts: reusable persona files you can apply to a changeset
  • Explicitly refuses to generate personas from an empty or sparse codebase

Limitations

  • Does not perform the review itself; you must explicitly ask Claude to run the persona afterwards
  • Multi-step interactive workflow (project type declaration, findings confirmation, role selection) that requires the user to stay in the loop
  • Custom roles outside the menu have no built-in seed guidance, so output quality is less predictable
When to use it

Use it when you want to set up recurring, stack-aware persona-based code review for an existing project.

When not to use it

Avoid it on an empty or scaffolding-only repo, or when you need an immediate review without interactive setup.

Security analysis

Safe
Quality score78/100

The skill is a codebase-reading workflow that scans project files with Grep/Glob/Read and Bash for inspection, then generates markdown persona files with Write. All operations are local and legitimate; there is no obfuscation, no remote execution, no destruction, and no credential handling. The Bash and Write tools are powerful in principle but are constrained to reading the repo and writing documentation-style outputs, so overall risk is low.

Findings
  • •Declares Bash/Write and writes persona markdown files to disk (default personas/), but scope is confined to a project-local output directory
  • •No network calls, no curl|sh, no destructive commands, no secret/token handling or exfiltration
How this analysis is made

Examples

Generate personas for an iOS app
Generate review personas for this project — it's an iOS app built with SwiftUI. I want security, platform, and QA perspectives.
Security and performance focus
Create review personas for this Express API codebase, focusing on security and performance.
Full persona suite
Build review personas for this WordPress plugin. Scan the codebase first and then generate all the roles you offer.

name: persona-generator description: "Generate tailored review personas for any project. Reads the codebase, presents a role menu, and produces persona files with domain-specific expertise, review criteria, and voice." allowed-tools: "Bash Read Write Glob Grep"

Persona Generator

Generate review personas tailored to your project's tech stack. Each persona is a markdown file you can hand to Claude as a review lens — "run this persona over the changeset."

Note: This skill generates personas. To run them, tell Claude directly: "Run the security-reviewer.md persona over the changeset."

When to Use

  • The user says "generate personas", "create personas", or "build review personas"
  • The user asks for review personas for a project
  • The user wants to set up persona-based code review for a codebase

Workflow

Follow all five steps in order. Do not skip steps. Do not generate personas until Step 3 is complete.

Step 1 — User Declares Project Type

Ask the user:

"What kind of project is this? (e.g., iOS app, WordPress plugin, React SPA, Express API)"

Free-form answer. This declaration anchors all persona generation — use it alongside codebase findings to tailor content.

Step 2 — Read the Codebase

First, verify the working directory is a project root. Check for source files, a package manifest (Package.swift, composer.json, package.json, etc.), or a git repository. If none of these exist, ask the user to confirm they're in the right directory.

From the project root, scan the codebase to understand the tech stack:

  1. Languages and runtimes — Check file extensions, package manifests, config files.
  2. Frameworks — Read dependency files (Package.swift, composer.json, package.json, requirements.txt, etc.) and identify major frameworks.
  3. Testing — Find test directories and existing test files to identify the testing framework and conventions.
  4. Architecture — Scan directory structure and key files to understand the architectural pattern.
  5. Key tools — Check for CI configs, build tools, linters, formatters.
  6. Conventions — Read a few representative source files to identify coding conventions, naming patterns, and style.

Present findings to the user for confirmation:

Here's what I found in the codebase:

  • Language/Runtime: [e.g., Swift 6, PHP 8.1]
  • Frameworks: [e.g., SwiftUI, UIKit, WordPress Plugin API]
  • Testing: [e.g., XCTest, PHPUnit + Brain Monkey]
  • Architecture: [e.g., MVVM, MVC with service layer]
  • Key tools: [e.g., SPM, Composer, GitHub Actions]
  • Conventions: [e.g., strict concurrency, WPCS, PSR-4]

Does this look right? Anything to add or correct?

Wait for the user to confirm or correct before proceeding.

If the codebase is sparse (few source files, no dependencies, scaffolding only), tell the user:

"This codebase doesn't have enough code for me to generate meaningful personas. Run this skill again once the repo is more populated."

Stop here. Do not generate personas from an empty codebase.

Step 3 — Present Role Menu & User Selects

Read references/roles.md and present the available roles:

Available persona roles:

  1. Security Reviewer — audits for vulnerabilities, data exposure, and unsafe patterns
  2. Platform Engineer — evaluates architecture, idioms, performance, and testability
  3. UI/UX Designer — reviews interface design, accessibility, and platform guidelines
  4. QA Engineer — evaluates test coverage, edge cases, and regression risk
  5. Performance Engineer — focuses on profiling, optimization, and resource efficiency

Which roles do you want? (e.g., "1, 2, 3" or "all" or "security and UI")

If the user requests a role not in the menu, generate it as a custom role. Tell the user:

"That's a custom role — I don't have built-in seed guidance for it, so review the output more carefully."

After role selection, ask two follow-up questions:

Specificity:

"Want me to include project-specific references (actual class names, UI patterns, tools from your codebase) or keep these generic to the stack?"

If the user wants project-specific references but the codebase doesn't have enough distinctive content to meaningfully personalize, say so and fall back to stack-generic content.

Sections:

"The default sections are Role, Expertise, Review Criteria, and Voice. Want to add or remove any sections?"

Wait for the user to answer both questions before proceeding to Step 4.

Step 4 — Generate Personas

For each selected role, generate a persona file:

  • Use references/persona-template.md as the structural guide
  • Use references/roles.md for role-specific seed content (focus areas, review lens, voice guidance)
  • Tailor expertise, review criteria, and voice to the tech stack and codebase findings from Step 2

Content guidelines:

  • Expertise should include 8-12 stack-specific skills. Not "database security" but "$wpdb->prepare() and parameterized queries" for WordPress, or "Keychain Services and App Transport Security" for iOS.
  • Review Criteria must be actionable. Each item is something the persona evaluates when looking at code, not a vague principle.
  • Voice must describe how the persona communicates, not what it knows. Include what references it cites, how it formats findings, and its communication style.

If the user requested project-specific references, weave in actual class names, patterns, and tools found in the codebase. If they chose generic, keep content at the stack level.

If the user requested custom sections, add them after Voice.

Security Reviewer note: The generated security reviewer persona must include this note in its Voice section: "This persona is an assistive review tool. For production systems handling sensitive data, authentication, or payments, validate findings with a human security review."

Ask where to save:

"Where should I save the personas? Default: personas/ in the project root."

If the directory doesn't exist, create it.

Check for existing personas: If the target directory already contains persona files, list them and ask:

"I found existing personas in personas/: [list]. Want me to overwrite all of them, or just generate specific ones?"

Only write the files the user confirms.

File naming: Use lowercase kebab-case matching the role name (e.g., security-reviewer.md, platform-engineer.md, ui-ux-designer.md, qa-engineer.md, performance-engineer.md). Custom roles follow the same pattern.

Step 5 — Present & Confirm

Show the user what was generated: list of files with a preview of each (the Role section):

Generated personas in personas/:

  • security-reviewer.md — [First sentence of Role section]
  • platform-engineer.md — [First sentence of Role section]
  • ui-ux-designer.md — [First sentence of Role section]

Want to revise any of these, or are they good?

If the user requests revisions, make them and show the updated preview. Repeat until the user approves.

After approval:

"Personas saved to personas/. To use them, tell Claude: 'Run the security-reviewer.md persona over the changeset.'"

From the file to a first result

The download contains exactly the content displayed on this page. This copy may differ from the latest version in the source repository.

View the source and supporting files

This download contains SKILL.md only. Any scripts, references, templates and dependencies are not included. Get the required files from the source, preserving their directory structure, before continuing.

1. Download the file

Download SKILL.md

From the root of a test project, save the file as .skill-downloads/persona-generator-mrwakaiw/SKILL.md, or use the command below. This folder is for inspecting the skill before installation.

Command for macOS/Linux or WSL: requires a POSIX shell and curl. On Windows without WSL, use the download button and create the folder manually.

mkdir -p '.skill-downloads/persona-generator-mrwakaiw' && curl --fail --show-error --location 'https://skills-guides.com/api/skills/persona-generator-mrwakaiw/download' --output '.skill-downloads/persona-generator-mrwakaiw/SKILL.md.part' && mv '.skill-downloads/persona-generator-mrwakaiw/SKILL.md.part' '.skill-downloads/persona-generator-mrwakaiw/SKILL.md'

2. Check prerequisites

Read SKILL.md before using it. If it references scripts or external services, inspect them and prepare the required access in your test environment.

Tools declared by the author (check availability in your agent): Bash Read Write Glob Grep

No recognized prerequisites section in this file. This does not mean it is self-contained: libraries, tools and access requirements may be mentioned elsewhere in the content or source repository.

The platforms on this page are catalog indications. The paths below follow tool documentation; they do not prove this skill has been run in those tools. No minimum tool version is confirmed for this listing.

3. Load the skill

After inspection, copy the complete folder into your project using one of the commands below, then open a new tool session. The command stops if a folder with the same name already exists: compare the files before replacing it.

Claude Code
test -f '.skill-downloads/persona-generator-mrwakaiw/SKILL.md' && test -s '.skill-downloads/persona-generator-mrwakaiw/SKILL.md' && mkdir -p '.claude/skills' && test ! -e '.claude/skills/persona-generator' && cp -R '.skill-downloads/persona-generator-mrwakaiw' '.claude/skills/persona-generator'

OpenSkills

OpenSkills must already be installed. The local install/read procedure is checked with OpenSkills 1.5.0, which requires Node.js ≥ 20.6.0. Catalog slugs are not resolved automatically.

Official OpenSkills installation
openskills install './.skill-downloads/persona-generator-mrwakaiw'

openskills read 'persona-generator-mrwakaiw'

Reading should display the installed file content and its base directory. It does not run the skill or install its dependencies.

For an agent that reads AGENTS.md, then sync the index. This command changes AGENTS.md: review the diff before using it.

openskills sync

Manual method or another tool: attach SKILL.md to a conversation, or open it in an agent that can read files. Include any required supporting resources. No universal folder is recognized by every tool.

4. Try a first task

Use a short, synthetic input: an API endpoint, a small document or an interface brief. In a new conversation, attach the file or give the agent its path, then adapt this request:

Read .skill-downloads/persona-generator-mrwakaiw/SKILL.md and apply its instructions to this example: [describe a test input]. Identify any missing files, dependencies or access. Produce the result requested by the skill without inventing data or running unreviewed scripts.

Expected result: output matching the skill instructions, or a precise list of missing prerequisites. A read confirmation alone does not validate a first use. If an external resource is unavailable, that path remains untested.

Related skills