Guide · Agent skill
How to use the Customize Product Counsel Profile agent skill
Updated · written with AI from the project's own documentation and checked against our catalogue data
This guide explains how to install and use the customize agent skill from anthropics/claude-for-legal, which lets you change one part of a product counsel profile without re-running the full setup. It is written for readers who have never used agent skills before. By the end you will know what the skill does, how to add it to your agent, how to trigger it with a plain request, and what it needs and does not do.
Customize Product Counsel Profileanthropics/claude-for-legal · Apache-2.0 Download the skill Quick start
- 1Confirm setup is done. Run /product-legal:cold-start-interview first if you have not already. The customize skill edits an existing profile and stops if setup is missing.
- 2Install the skill. Download the ZIP from this page and unzip it into ~/.claude/skills/ or a project's .claude/skills/ folder, or upload it under Settings → Capabilities → Skills in the Claude apps.
- 3Start a customize request. Type /product-legal:customize or ask in plain words, such as "update my escalation contacts".
- 4Pick one section. Review the grouped map of your profile and choose the single section you want to change.
- 5Review the current value. The skill shows the existing value so you can see exactly what will change.
- 6Give and confirm the new value. Provide the new value and read the downstream explanation and any inconsistency warnings before confirming.
- 7Check the saved change. Once confirmed, the skill writes the change to the config and states what the next outputs will reflect.
What is the customize skill and what does it do?
An agent skill is a folder containing a SKILL.md file that gives an agent a set of instructions for a specific task. This skill, named customize, is for legal or product counsel who already completed the cold-start setup and want to adjust a single setting. It covers risk calibration, escalation contacts, launch review categories, marketing claims posture, and workspace paths.
The skill first reads your current config, then shows you a map of the profile grouped by area with the current value of each. It asks what you want to change, shows the old value, confirms the new one, and saves it to a configuration file. It also warns you if a change makes the profile inconsistent or weakens safeguards that the wider plugin depends on.
What do you need before installing?
The skill requires Claude Code or a compatible agent with the claude-for-legal product-legal plugin installed. It also requires that you have already completed /product-legal:cold-start-interview, because that interview creates the config files this skill reads and writes.
If you have not run setup, the skill stops and tells you to run the cold-start interview first. It is not meant for starting a profile from scratch.
- Claude Code, or a compatible agent
- The claude-for-legal product-legal plugin installed
- Completed /product-legal:cold-start-interview
How do you install the customize skill?
Download the ZIP file for this skill from this page and unzip it. The skill ships as a single file, SKILL.md. To install it for yourself across projects, place the unzipped folder in ~/.claude/skills/. To install it for one project only, place it in that project's .claude/skills/ folder.
If you use the Claude apps instead, upload the skill under Settings → Capabilities → Skills.
- Personal install: unzip into ~/.claude/skills/
- Project install: unzip into the project's .claude/skills/
- Claude apps: Settings → Capabilities → Skills, then upload
How do you trigger the skill with a natural request?
You can invoke the skill with the command /product-legal:customize, optionally followed by a section name or a description of what you want to change. You can also ask in plain language. The skill is triggered by requests such as "change my [thing]", "update my profile", "edit my framework", "retune my calibration", or "customize".
Good requests name one thing. For example, you might say you want to change your escalation contact, or that you want to retune your risk calibration. The skill works one change at a time and will not re-ask the whole interview.
- "Update my escalation contacts"
- "Retune my calibration" or "edit my framework"
- /product-legal:customize marketing claims
/product-legal:customizeWhat happens during an edit?
The skill shows the current value of the section you choose, asks for the new value, explains what changes downstream, and asks you to confirm before writing it to the config. Existing outputs are not rewritten automatically. For example, after a risk calibration change, the skill notes that existing reviews stay as written and that you can re-run a review if you want the new posture applied.
For changes to shared company details such as company name, industry, jurisdictions, practice setting, or stage, the skill writes to company-profile.md. It then notes that the change affects all 12 plugins that read your jurisdiction footprint.
- Changes to a single section are written to the product-legal config
- Shared company details are written to company-profile.md and affect all 12 plugins
- Removing a review category is replaced by marking it as not in scope and naming who handles it
What are the safety notes?
The skill does not run scripts and does not use the network. It reads and writes configuration files only. It does not assess any product or claim, and it does not give legal advice.
The skill is designed to protect safeguards. It will not delete a section. It flags when a change makes the profile inconsistent, for example turning on AI-feature claims scrutiny without AI policy commitments set in /ai-governance-legal, or pairing a fast review SLA with a rule that every launch needs GC sign-off. It also discourages removing the review flag, source attribution tags, or verify tags on cited regulations, and it treats the substantiation requirement on claims as the thing /marketing-claims-review exists for.
- No scripts are run
- No network use
- Configuration files only; no product or claim assessment
- Substantiation requirements, source tags, and verify tags are kept
What people use it for
- Updating the escalation chain when your manager, GC, or risk committee changes
- Tightening a risk calibration so a pattern moves from fine to needs a real look
- Adding a new launch-review category for a regulation your team now faces
- Tightening the marketing claims posture so more language is flagged for substantiation
- Updating jurisdictions or practice setting in the shared company profile after a change in your footprint
Frequently asked questions
Is the customize skill free?
The source does not state pricing or licence terms for this skill beyond its licence, which is Apache-2.0. Check the repository for details on the claude-for-legal project.
Do I need to run setup before using customize?
Yes. The skill reads config files created by /product-legal:cold-start-interview. If setup has not been run or the config still contains placeholder values, the skill tells you to run the interview first.
Where do I install the customize skill?
Unzip the downloaded ZIP into ~/.claude/skills/ for personal use or into a project's .claude/skills/ folder for one project. In the Claude apps, upload it under Settings → Capabilities → Skills.
Does customize give legal advice or check my product?
No. It only edits your configuration. It does not assess any product, feature, or marketing claim, and it is not a source of legal advice.
Can I remove a review category or safeguard with customize?
You cannot delete a section. To remove a review category, the skill marks it as not in scope and names the plugin or team that picks it up. It discourages removing substantiation requirements, source tags, or verify tags because they are load-bearing.
Does customize run scripts or use the internet?
The source states that this skill runs no scripts and uses no network. It reads and writes configuration files only.
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 customize if you already completed the product-legal cold-start setup and need to change one setting, such as escalation contacts, calibration, or a review category, without hand-editing config files. Pick something else if you have not run setup yet, or if you need someone to assess a product, feature, or claim, since this skill only edits configuration.
Download the skill