make-req
Skill1 starsChanged 8 months agoMITView on GitHub
Decompose a task, plan, roadmap, or feature description into a structured set of requirements grouped for independent planning. Use when user says "make req", "make requirements", "decompose requirements", "extract requirements", or wants to break down a task into what needs to be true.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- TalonT-Org/AutoSkillitword for word, first seen May 8, 2026
Other versions1 project uses a version of this.
Word for word · 1
- TalonT-Org/AutoSkillit★ 5100% same
Edited · 0
None found.
Wrote their own version · 0
None found.
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 repository59 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- arch-lens-c4-containerSkill
- arch-lens-concurrencySkill
- arch-lens-data-lineageSkill
- arch-lens-deploymentSkill
- arch-lens-developmentSkill
- arch-lens-error-resilienceSkill
- arch-lens-module-dependencySkill
- arch-lens-operationalSkill
- arch-lens-process-flowSkill
- arch-lens-repository-accessSkill
- arch-lens-scenariosSkill
- arch-lens-securitySkill
- arch-lens-state-lifecycleSkill
- audit-archSkill
- audit-bugsSkill
- audit-cohesionSkill
- audit-defense-standardsSkill
- audit-frictionSkill
- audit-implSkill
- audit-testsSkill
- design-guardsSkill
- dry-walkthroughSkill
- elaborate-phaseSkill
- exp-lens-benchmark-representativenessSkill
- exp-lens-causal-assumptionsSkill
- exp-lens-comparator-constructionSkill
- exp-lens-error-budgetSkill
- exp-lens-estimand-claritySkill
- exp-lens-exploratory-confirmatorySkill
- exp-lens-fair-comparisonSkill
- exp-lens-governance-riskSkill
- exp-lens-iterative-learningSkill
- exp-lens-measurement-validitySkill
- exp-lens-pipeline-integritySkill
- exp-lens-randomization-blockingSkill
- exp-lens-reproducibility-artifactsSkill
- exp-lens-sensitivity-robustnessSkill
- exp-lens-severity-testingSkill
- exp-lens-unit-interferenceSkill
- exp-lens-validity-threatsSkill
- exp-lens-variance-stabilitySkill
- id-slopSkill
- implement-worktreeSkill
- investigateSkill
- make-arch-diagSkill
- make-experiment-diagSkill
- make-groupsSkill
- make-planSkill
- make-scenariosSkill
- mermaidSkill
- plan-arcSkill
- rectifySkill
- review-approachSkill
- update-architectureSkill
- update-reqsSkill
- update-specsSkill
- verify-diagSkill
- verify-proposed-diagSkill
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
- Very specific: line 34
Create files outside temp/make-req/ directory
- Distinctive: line 37
Write requirements that can only be verified by reading source code
- Distinctive: line 40
Use subagents to understand the source material and relevant codebase context
- Very specific: line 44
Write to temp/make-req/ directory
- Distinctive: line 52
If the input references existing systems or codebases, launch parallel Explore subagents to understand:
- Distinctive: line 89
Keep context factual. Do not suggest solutions.
- Distinctive: line 142
No requirement tells the reader what to build or how to build it
- Distinctive: line 143
Every requirement is a testable condition
- Distinctive: line 146
No group depends on understanding another group's requirements to be plannable (cross-references to other groups are allowed for ordering, …
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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
Requirements Decomposition Skill
Decompose a task, plan, roadmap, or feature description into a structured set of requirements. Requirements state what must be true when the work is done — never how to achieve it.
When to Use
- User says "make req", "make requirements", "decompose requirements"
- User has a task description, plan, roadmap, or feature brief and wants requirements extracted
- User wants to break a large effort into independently plannable groups
Core Principles
- Requirements are acceptance criteria, not instructions. Each requirement states a condition that must hold. It does not prescribe an approach, technology, pattern, or implementation step.
- Group by co-implementation. Requirements that would naturally be delivered together belong in the same group. Each group should be independently plannable — someone could take a single group and produce a complete plan from it without needing to implement other groups first (though groups may have ordering preferences).
- Describe observable outcomes. A requirement should be verifiable by observing the system's behavior, interfaces, or data — not by inspecting its internals.
- Source material is unverified input. Treat factual claims in the input (plans, scenarios, task descriptions, reports) as hypotheses about the codebase. Before incorporating any claim about how the system works into your output, verify it against subagent findings. When source material and codebase disagree, the codebase is authoritative.
Critical Constraints
NEVER:
- Modify any source code files
- Create files outside
temp/make-req/directory - Prescribe implementation approaches or libraries in requirements
- Include implementation steps disguised as requirements ("Refactor X to use Y" is an instruction, not a requirement)
- Write requirements that can only be verified by reading source code
ALWAYS:
- Use subagents to understand the source material and relevant codebase context
- Group requirements by what would be implemented together
- Provide background and context for each group
- State requirements as verifiable conditions
- Write to
temp/make-req/directory
Workflow
Step 1: Understand the Source Material
Identify the input: a task description, plan document, roadmap, conversation context, or file reference. Read it fully.
If the input references existing systems or codebases, launch parallel Explore subagents to understand:
- What exists today that the requirements relate to
- Current capabilities and boundaries
- Domain terminology and concepts used in the source material
Step 2: Extract Raw Requirements
From the source material and codebase understanding, identify every distinct thing that must be true when the work is complete. At this stage, don't group — just capture.
For each candidate requirement, test it:
- Does it state a condition, not an action?
- Can it be verified without reading source code?
- Is it a single requirement, not multiple joined by "and"?
- Does it avoid naming specific technologies, patterns, or implementation approaches?
Rewrite any that fail these tests. Discard any that are purely implementation concerns with no observable outcome.
Step 3: Identify Groups
Cluster requirements into groups based on what would be delivered together. Consider:
- Which requirements share a domain or functional area?
- Which requirements have mutual dependencies (one is meaningless without the other)?
- Which requirements could be handed to a separate team or planning effort?
Name each group with a short descriptive label. Order groups by suggested implementation sequence where dependencies exist — note these as ordering preferences, not hard constraints.
Step 4: Add Context
For each group, write a brief background section covering:
- What this group of requirements pertains to
- Why these requirements exist (the problem or need they address)
- What a reader needs to know to plan an implementation
- Any relevant constraints, boundaries, or dependencies on other groups
Keep context factual. Do not suggest solutions.
Step 5: Write the Requirements Document
Save to: temp/make-req/requirements_{topic}_{YYYY-MM-DD_HHMMSS}.md
# Requirements: {Topic}
**Date:** {YYYY-MM-DD}
**Source:** {What was decomposed — task name, plan reference, or brief description}
## Overview
{One paragraph: what these requirements collectively describe and their intended use}
---
## {Group 1 Name}
### Background
{What this group pertains to, why it matters, relevant context}
### Requirements
- **REQ-{GRP}-001:** {Requirement statement}
- **REQ-{GRP}-002:** {Requirement statement}
- ...
{If this group depends on or relates to another group, note it briefly after the requirements list.}
---
## {Group 2 Name}
### Background
{...}
### Requirements
- **REQ-{GRP}-001:** {Requirement statement}
- ...
---
{Repeat for each group}Numbering: {GRP} is a short uppercase abbreviation of the group name (e.g., AUTH, DATA, UI). Numbers are sequential per group starting at 001.
Output Quality Checks
Before finalizing, verify:
- No requirement tells the reader what to build or how to build it
- Every requirement is a testable condition
- Groups are cohesive — moving a requirement to a different group would feel wrong
- A reader unfamiliar with the codebase could understand what each group demands from the background section alone
- No group depends on understanding another group's requirements to be plannable (cross-references to other groups are allowed for ordering, not comprehension)
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.

