Guide · Agent skill
How to use the Wiggins Narrative Profile agent skill
Updated · written with AI from the project's own documentation and checked against our catalogue data
The Wiggins Narrative Profile is an agent skill, a set of instructions you add to an AI agent so it behaves in a specific way. This one makes the agent focus on the meaning of a decision: why it was made, what the framing is, and how to explain it to different readers. This guide shows you what the skill does, how to install it, how to trigger it with plain requests, and what to watch for.
Quick start
- 1Download the ZIP. Get the skill ZIP from this page. It contains SKILL.md.
- 2Choose where to install. Use ~/.claude/skills/ for personal use, or the project's .claude/skills/ folder for one project.
- 3Unzip the file. Extract the ZIP into the folder you chose so the skill files sit inside it.
- 4Or upload it in the Claude apps. In the Claude apps, go to Settings → Capabilities → Skills and upload the ZIP.
- 5Ask for a narrative task. Describe a decision, PR or design note, and name the audience who will read it.
What the Wiggins skill makes the agent do
The skill gives the agent a persona called Wiggins, described as a narrative and synthesis partner. Wiggins focuses on meaning, framing and communication quality rather than on writing code or carrying out implementation work.
By default, Wiggins challenges the reasoning behind a decision before it challenges the conclusion. It points out assumptions and framing choices, keeps the tone calm and plain, and offers alternative phrasings for different audiences such as engineers, leaders or partners.
- Explain why a decision was made
- Write or refine PR descriptions and design notes
- Translate technical details for non-technical readers
- Synthesize tradeoffs across several inputs
- Test whether your reasoning and story actually hold together
Install the Wiggins skill from the ZIP or the Claude apps
Download the ZIP file for the skill from this page. The skill folder contains a single file, SKILL.md, so there is nothing else to set up.
You have two ways to install it. For a personal setup, unzip the download into your ~/.claude/skills/ folder. To limit it to one project, unzip it into that project's .claude/skills/ folder instead.
If you use the Claude apps, you can upload the skill instead. Go to Settings, then Capabilities, then Skills, and add the ZIP there.
- Personal use: unzip into ~/.claude/skills/
- Single project: unzip into the project's .claude/skills/
- Claude apps: Settings → Capabilities → Skills
How do you trigger the Wiggins skill with a request?
The skill is meant to be used when you ask for one of a few kinds of work. You do not need a special command. Describe the task in plain words and say what you want the output to do for its reader.
The source lists these situations as the ones where Wiggins should be used: explaining why a decision was made, writing or refining PR descriptions and design notes, translating technical details for non-technical readers, and synthesizing tradeoffs across multiple inputs.
- Explain why we chose a queue over direct API calls, for our operations lead.
- Rewrite this PR description so a product manager can follow it.
- Here are three design options and their tradeoffs. Synthesize them into a short note for the partner team.
- Challenge my reasoning for moving to this vendor. Where are the weak assumptions?
Examples of good requests and what to expect
Give the agent the decision, the audience and the material you have. Wiggins will tend to say what is known, what is inferred and what is uncertain, rather than presenting every claim with equal confidence. That helps you see which parts of your story still need support.
Expect the agent to ask what matters to the reader and to suggest more than one phrasing when the audience is mixed. It is designed to make explanations clearer without overstating the case, so it may trim claims you wanted to make bigger.
- Ask for two versions of the same note: one for engineers, one for leadership.
- Ask it to name the tradeoffs and the consequences of the decision explicitly.
- Ask it to point out framing choices that could make a weak argument look stronger than it is.
What the skill does not do and what it needs
The skill lists nothing as a requirement. It does not run scripts and does not use the network. It is written for narrative work, so it does not include hands-on implementation or code change steps. The skill itself says those are better handled by a different profile called Quinn.
The same is true for fact-checking. The skill says evidence analysis belongs to a profile called Anitta, and it does not add verification steps of its own. The file describes how Wiggins hands off to Quinn and Anitta, but those two sibling skills are not part of this folder, so the handoffs will not work on their own.
- Requirements: none listed
- Runs scripts: no
- Uses network: no
- Not included: the Quinn and Anitta profiles referenced in handoffs
What people use it for
- Writing a PR description that a non-technical colleague can follow.
- Explaining to leaders or partners why a technical decision was made.
- Turning several design options and their tradeoffs into one clear note.
- Checking whether your reasoning and story hold together before you share them.
Frequently asked questions
Is the Wiggins agent skill free?
The source lists the skill as MIT licensed. It does not mention prices, so check the licence file and the terms of your own agent platform for any costs.
What do I need to install the Wiggins skill?
The skill lists no requirements. Download the ZIP, then unzip it into ~/.claude/skills/ or a project's .claude/skills/ folder, or upload it under Settings → Capabilities → Skills in the Claude apps.
Does the Wiggins skill run scripts or use the internet?
No. The skill runs no scripts and uses no network. It consists of a single SKILL.md instruction file.
Can Wiggins write code or fix bugs?
No. The skill is for framing and explanation. The skill says implementation work is better handled by a different profile, Quinn, which is not included in this folder.
How is Wiggins different from Anitta and Quinn?
Wiggins is interpretive and narrative-focused. Anitta is evidence-forward and investigative, and Quinn focuses on implementation and execution. Only the Wiggins file is included here.
Why doesn't the Wiggins skill work with the handoffs it describes?
The file references Quinn and Anitta, but those sibling skills are not part of this download. You can use Wiggins alone, but the coordinated workflow needs those other profiles.
New agent skills, repos and AI tools — every week
The best new skills and open-source projects, the fastest-growing repos, and one free prompt. Five minutes to read. Unsubscribe anytime.
Our verdict
Use this skill if you write PR descriptions, design notes or decision summaries that non-technical readers must follow, or if you want an agent to test whether your reasoning holds together. Pick something else if you need code changes, implementation help or evidence checking, since the skill itself points those tasks elsewhere and does not include the profiles that handle them.
Download the skill