From a Template
Fork a public alter-ego and feed it your data to make it your own.
Public alter-egos are templates the community has built and shared. They come with a personality, a stance, and a set of moves, but they have never seen your data. The moment you fork one and feed it your work, it stops being a template and becomes a private alter-ego of yours.
Forking does not pull your data into anyone else's workspace, and it does not pull the original author's data into yours. You inherit the shape, then you fill it.
When forking beats starting from scratch
Start from a template when:
- You want a disposition you do not naturally have. "The Patient Investor" forces longer time horizons into your thinking. "The Mentor" gently turns advice into questions. You are borrowing a posture you would have to write the brief for otherwise.
- You are new to alter-egos and want something working in fifteen minutes. The template gives you a draft on day one, even before you have fed it anything.
- The template author has already done the structural work. "The Strategist" comes with a prompt and a set of frameworks already loaded. You skip the blank page.
Start from scratch (see From scratch) when the surface is highly idiosyncratic to you or your firm, when no public template comes close, or when voice fidelity matters more than disposition.
A tour of what is in the catalog
The public catalog is community-built and grows over time. Today's most-forked templates fall into three rough buckets.
| Bucket | Examples | What you get |
|---|---|---|
| Dispositions | The Patient Investor, The Skeptic, The Optimist | A stance to lean against. Voice is light; the brief is doing the work. |
| Roles | The Mentor, The Strategist, The Editor | A working pattern around a kind of conversation. Comes with prompts and follow-up moves. |
| Surfaces | The Memo Writer, The Update Writer, The Cold Email | A specific artifact and a structure for producing it. Closest to a "useful on day one" template. |
Surfaces forks the fastest. Dispositions take the longest to make your own, because the gap between the template's stance and your data is wider.
Forking and personalising
Pick a template
Browse the public catalog and pick one whose surface or disposition is close to what you want. Read its brief and a couple of sample outputs before forking. If the sample outputs already feel wrong, do not fork. Move on.
Fork it into your workspace
Forking creates a private copy of the template's brief and configuration inside your workspace. The original stays public and untouched. Anything you do to your fork from this point on stays in your workspace.
Feed it your data
This is the step that turns the template into yours. Drop the source material that matches the template's surface into the fork's folder:
- For a memo template: your past memos and the calls behind them.
- For a mentor template: transcripts of times you have mentored someone, your own notes from being mentored, any writing you have done on the topic.
- For a cold email template: your sent folder filtered to cold outreach that actually got replies.
The template's brief stays in place. The data sits underneath, and the alter-ego starts pattern-matching against it on the next run.
Tune the brief, do not rewrite it
The author wrote the brief for a reason: it captures the disposition you forked for. Resist rewriting it from scratch. Instead, add a short personalisation block at the end:
- "Always close with a question that puts the next move on the other person."
- "Use the way I phrase things in the source material, even when it disagrees with the template's defaults."
- "Never use the word 'leverage' or 'synergy.'"
Small, specific overrides keep the template's value and bend it toward you.
Run it on a real task
Try it on a task you have a known good answer for, the same way you would test a from-scratch alter-ego. The gap between the template's draft and the answer you would have written is your tuning signal.
What changes when a fork becomes a private alter-ego
The moment a fork has your data in it, it is a private alter-ego in every meaningful sense: only you can use it, only your data feeds it, and it lives in your workspace. The "forked from" link stays around so you remember where it came from, but you are not sharing anything back upstream by default.
If the template author publishes an update to their brief later, your fork does not auto-update. You will see a notice, and you can choose to pull in the changes, merge them with your overrides, or ignore them.
A worked example
Someone wants a sharper draft of weekly investor updates, but writing the brief for a "good investor update" themselves is the work they are trying to avoid.
- They fork "The Update Writer," which already encodes a structure: highlights, lowlights, asks, metrics.
- They drop in their last six months of investor updates plus the Slack channel where the team debriefs each week.
- They append a five-line personalisation block: stay under 400 words, never put metrics before narrative, always end with two specific asks.
- The next Monday, the alter-ego produces a draft. It has the template's structure and a voice that is recognisably theirs.
That is the win-state for forking: the template did the structural work, the data did the voice work, and the brief is mostly the original author's plus a handful of overrides.