Claude, Expansion Update.
The expansion-update skill compares an existing employee handbook against a new state's requirements and flags every section that needs a change, leaving the actual rewrite to the reader.
Can Claude rewrite your employee handbook for a new state? Not quite. The expansion-update skill compares your existing handbook or policy against what the new state or scope actually requires, then flags every section that needs a change. Deciding what the new handbook should actually say remains a human decision.
What a Claude skill is in this context
A skill is a folder Claude reads before it works. This one is called expansion-update, and its SKILL.md file holds the full instruction set in plain language, with no hidden logic. Claude reads the file, then acts on it. The file is the program that defines the behavior, not an open-ended interpretation of the request.
How the pipeline runs
The instructions sit in a step section, and Claude reads each step in order, then runs it. The process is linear: no branching unless a specific step says to branch. That predictability is deliberate. It means the same process applies whether the handbook is moving into one new state or several.
The specific constraint this skill runs
The task is narrow and specific: compare the existing handbook or policy against what the new state or scope requires, then flag every section that qualifies as needing a change. Staying inside that checklist means the flagged list holds the same shape every time it runs. Same checklist in, same flagged sections out, consistently.
What this does and doesn't decide
The skill's output is a list of sections that need attention because the new jurisdiction's requirements differ from what the current handbook says. It does not draft the replacement language, and it does not decide how the company wants to handle a given requirement where multiple compliant approaches exist. That judgment, what the new handbook actually says, stays with the person running the skill.
Using it on your own handbook
The practical move is to give Claude your current employee handbook, tell it you're expanding into a new state, and ask it to read the expansion-update skill and explain exactly what it needs from you before running any check. Then it flags what needs updating and explains why. That sequencing, asking what it needs before it checks anything, mirrors the same discipline shown throughout this skill family: the process is fixed, and the tool tells you its own inputs before acting on them.
Key takeaways
- The expansion-update skill flags handbook sections that need changes for a new state; it does not rewrite the handbook.
- The skill's behavior comes from a fixed instruction file, read and followed step by step, with no branching unless specified.
- The task is narrowly scoped: compare against the new jurisdiction's requirements, flag qualifying sections.
- The same checklist produces the same flagged sections every time it runs on the same inputs.
- Deciding what the updated handbook should actually say remains the user's responsibility.
Who this is for
HR and compliance teams managing an employee handbook across multiple states, who want a consistent first pass at identifying what needs review before drafting updated language themselves.
Full transcript(auto-generated, with timestamps)
[0:00]Someone wondered if we're expanding into a new state, can Claude just rewrite our handbook for it? Not quite. It flags what needs to change. Here's what actually happens when you run it. A skill is a folder Claude reads before it works. This one is expansion update. The skill MD file holds the full instruction set in plain language, no hidden logic. Claude reads it then acts. The file is the program. The pipeline sits in the step section. Claude reads each step in order then runs it. Linear, no branching unless a step says so. The constraint is specific. Compare the existing handbook or policy against what the new state or scope requires, then
[0:36]Flag every section that needs a change. Stay inside that checklist and the flagged list holds its shape every time. Same checklist in, the same flagged sections out every time. Deciding what the new handbook actually says is still yours. Your turn. Here's the prompt. Read it with me. We're expanding into a new state and I have our current employee handbook. Read the expansion update skill. Tell me exactly what you need from me before you run any check, then flag what needs updating in the handbook and tell me why. Claude expansion update. Lay 'em in for bear.
