persona-brief
Skill1 starsChanged 42 days agoMITView on GitHub
Build a short reader brief before drafting — who the reader is, what they already know, what they need defined, what order the content must come in, and what they should do afterward. Use before /persona-write when the reader may not share the author's context, or on its own to check a draft against its reader.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- cal1888/claude-code-skillstheir own version, first seen Jul 18, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- cal1888/claude-code-skills★ 1same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository8 other files it gives its agents.
- persona-copySkill
- persona-createSkill
- persona-listSkill
- persona-localizeSkill
- persona-researchSkill
- persona-reviewSkill
- persona-wordcheckSkill
- persona-writeSkill
The file
We split this skill into the rules it gives an agent
9 rules. About 0 are common, 7 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈7
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 12
Use it for documentation, requirements, reports, PR descriptions, explanations, handoffs, and any text where the reader does not share the …
- Distinctive: line 25
If the reader or the wanted action is unclear and changes the result, ask one question.
- Distinctive: line 29
Write these seven fields, one to three lines each.
- Distinctive: line 41
Read the source and the stated reader.
- Distinctive: line 44
After drafting, run the reader check below on the result.
- Distinctive: line 48
Read the draft as the reader from field 1, with only field 2 in hand.
- Very specific: line 64
Run before /persona-write for new drafts and rewrites.
- Very specific: line 65
Run after /persona-review when the text is correct and in voice but the reader still has to guess.
- Distinctive: line 66
Not a substitute for fact-checking.
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Persona Brief
Purpose
A persona decides how the writer sounds. A reader brief decides what the reader needs in order to follow and act. This skill produces that brief, then hands it to /persona-write (or checks an existing draft against it).
Use it for documentation, requirements, reports, PR descriptions, explanations, handoffs, and any text where the reader does not share the author's context.
The writer this skill makes
The writer is someone who sits in the reader's seat before drafting. They know what they know, and they can tell which parts of it the reader does not have. They put context ahead of the detail that depends on it, define a term where it first earns its place, and finish with the reader able to act without asking the author a question.
Inputs
- source material (notes, code, data, a draft)
- the reader, even if only "a new teammate" or "the approver"
- the purpose and the action wanted from the reader
- stakes: what happens if the reader misunderstands
If the reader or the wanted action is unclear and changes the result, ask one question. Otherwise infer and say what was inferred.
The brief
Write these seven fields, one to three lines each.
- Reader — role, and how much attention they will give it.
- Already knows — context the reader brings. This is what can be left out.
- Does not know — context only the author has: terms, history, decisions, constraints. This is what must be supplied.
- Order — the sequence in which the reader needs things, each item placed after what it depends on.
- Action — what the reader should do, decide, or believe afterward.
- Stakes and register — how careful, how long, how formal.
- Impression — what the reader should think of the author or the text after reading (for example: credible, modest, direct). The cold-reader pass compares the reader's report with this.
Workflow
- Read the source and the stated reader.
- Fill the brief. Mark each line as given or inferred.
- Hand the brief to
/persona-writeas its audience and purpose input, alongside the chosen persona. The persona sets the voice; the brief sets the content and its order. - After drafting, run the reader check below on the result.
Reader check
Read the draft as the reader from field 1, with only field 2 in hand.
- Does every term appear after its definition, or with it?
- Does each detail come after the context it depends on?
- Can the reader state the action in field 5 without contacting the author?
- Is the length right for the stakes in field 6?
- Which claims are facts from the source, and which are the author's interpretation? Is that visible?
Fix gaps by adding the missing context at the point it is needed, not in a preface.
Output
Return the brief, then the draft or the fixes. Keep the brief short enough to read in under a minute.
Pairing
- Run before
/persona-writefor new drafts and rewrites. - Run after
/persona-reviewwhen the text is correct and in voice but the reader still has to guess. - Not a substitute for fact-checking; it works from the source material supplied.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

