prompt-brief-builder
Skill1 starsChanged 5 months agoMITView on GitHub
Convert rough task descriptions into structured prompt briefs. Use this skill when the user wants to create a high-quality prompt but only has a vague idea, business task, assistant concept, automation idea, or desired output.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- ChromiteCr/Skillstheir own version, first seen Oct 17, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- ChromiteCr/Skills★ 5same 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 repository9 other files it gives its agents.
- prompt-xray-skillsSkill
- prompt-auditorSkill
- system-prompt-architectSkill
- agent-instruction-writerSkill
- rag-answer-policySkill
- structured-output-designerSkill
- prompt-test-generatorSkill
- prompt-debuggerSkill
- prompt-to-product-specSkill
The file
We split this skill into the rules it gives an agent
9 rules. About 0 are common, 9 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈9
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 10
Use this skill to transform a vague request into a structured prompt brief.
- Distinctive: line 18
Use this skill when the user says things like:
- Distinctive: line 21
create a prompt for this task.
- Distinctive: line 31
Do not start by writing the final prompt.
- Distinctive: line 71
Avoid generic roles like:
- Distinctive: line 121
Read the user input.
- Distinctive: line 126
Run a final quality check.
- Distinctive: line 157
If the output will be consumed by software, recommend structured JSON.
- Distinctive: line 168
does not invent missing facts.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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
Prompt Brief Builder Skill
Purpose
Use this skill to transform a vague request into a structured prompt brief.
A prompt brief is not the final prompt.
It is the specification that should exist before writing a reliable prompt.
When to use
Use this skill when the user says things like:
- help me write a prompt;
- create a prompt for this task;
- I need an assistant that does this;
- turn this idea into a prompt;
- make a system prompt;
- prepare instructions for an AI agent;
- I want the model to generate better results;
- I do not know how to explain the task to an LLM.
Core principle
Do not start by writing the final prompt.
First, clarify the task structure.
A good prompt brief should define:
- Goal
- User
- Model role
- Context
- Inputs
- Workflow
- Constraints
- Output format
- Quality criteria
- Failure behavior
- Test cases
Procedure
1. Extract the task
Restate the task in one sentence:
The model should help [target user] achieve [goal] using [input/context] and return [output].2. Identify the user and audience
Clarify:
- who will use the output;
- who the output is for;
- whether the output should be technical, business, educational, creative, operational, or customer-facing.
3. Identify the model role
Define the model's role only if it improves behavior.
Avoid generic roles like:
- "You are an expert";
- "You are the best";
- "You are a genius".
Prefer specific roles:
- "You are a B2B sales research assistant";
- "You are a product-specification assistant";
- "You are a support-answer policy engine";
- "You are a coding-agent reviewer".
4. Define input data
List the data the model will receive:
- user text;
- files;
- website content;
- CRM record;
- customer message;
- code;
- database fields;
- transcript;
- previous conversation;
- retrieved documents.
For each input, define whether it is required or optional.
5. Define source of truth
Specify what the model should rely on:
- provided context only;
- uploaded files;
- retrieved documents;
- external search;
- internal business rules;
- user-provided facts;
- model knowledge.
Also define what the model must not invent.
6. Define workflow
Break the task into steps.
Example:
- Read the user input.
- Extract key facts.
- Identify missing information.
- Decide whether there is enough context.
- Generate the output.
- Run a final quality check.
7. Define constraints
Specify:
- tone;
- language;
- length;
- formatting;
- forbidden content;
- required sections;
- business rules;
- privacy limits;
- uncertainty behavior.
8. Define output format
Choose the format:
- plain text;
- Markdown;
- table;
- JSON;
- structured report;
- checklist;
- email draft;
- product spec;
- code block;
- schema-compliant object.
If the output will be consumed by software, recommend structured JSON.
9. Define quality criteria
List what a good result must satisfy:
- accurate;
- grounded in input;
- complete;
- concise;
- follows format;
- does not invent missing facts;
- handles edge cases;
- useful for the next action.
10. Define failure behavior
Specify what the model should do when:
- input is incomplete;
- sources conflict;
- the task is ambiguous;
- the requested output is impossible;
- there is not enough evidence;
- the user asks for unsupported claims.
11. Create test cases
Generate:
- normal case;
- edge case;
- missing-context case;
- ambiguous case;
- bad-input case.
Output format
Return the result as:
# Prompt Brief
## One-sentence task
The model should...
## Use Case
- User:
- Audience:
- Business goal:
- Output goal:
## Model Role
The model should act as...
## Inputs
| Input | Required | Notes |
|---|---:|---|
## Source of Truth
The model should rely on:
The model must not invent:
## Workflow
1.
2.
3.
4.
5.
## Constraints
- Language:
- Tone:
- Length:
- Format:
- Forbidden actions:
- Required behavior:
## Output Format
Define the exact output structure here.
## Quality Criteria
- [ ] Criterion
- [ ] Criterion
- [ ] Criterion
## Failure Behavior
When information is missing, the model should...
When sources conflict, the model should...
When the request is ambiguous, the model should...
## Test Cases
### Normal case
Input:
Expected output:
### Edge case
Input:
Expected output:
### Missing-context case
Input:
Expected output:
## Next Step
After creating the brief, use it to write the final prompt or pass it into another skill such as `system-prompt-architect`, `prompt-auditor`, or `prompt-test-generator`.Quality check
Before finalizing, verify:
- the task is specific;
- the source of truth is explicit;
- inputs are defined;
- the output format is clear;
- missing-data behavior is defined;
- the prompt brief can be used to write a real prompt.
Boundaries
This skill creates a prompt brief. It does not guarantee that the final prompt will work without testing.
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.

