njpattennjpatten/claude-skills/project-instructions-generator/SKILL.md

project-instructions-generator

Skill0 starsChanged 2 months agoMITView on GitHub

Generates Claude Project instructions for any use case including client engagements, product development, operational domains, or research projects. Produces token-efficient behavioral instructions that define how to operate. Triggers on 'create project instructions for', 'set up a Claude Project for', 'I need a copilot for', or 'help me configure a Claude Project'.

In this repository6 other files it gives its agents.
- content-repurposerSkill
- daily-check-inSkill
- linkedin-post-generatorSkill
- meeting-synthesizerSkill
- okr-managerSkill
- pre-call-prepSkill

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 4 are distinctive, and 0 are very specific to this project.

≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
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:
Create token-efficient behavioral instructions that define HOW a Claude Project should operate.
line 17
Distinctive:
All instructions follow the same structural pattern:
line 30
Distinctive:
Ask clarifying questions first: if inputs are incomplete
line 83
Distinctive:
Before finalizing instructions, verify:
line 90

≈ “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.

Keep what fits and drop the rest. Your version, ready for your agents.

Files

SKILL.md

Project Instructions Generator

Generate Claude Project instructions for any use case: client engagements, product development, operational domains, or research projects.

Purpose

Create token-efficient behavioral instructions that define HOW a Claude Project should operate. Instructions are distinct from memory (facts/state) and project files (reference materials).

Use Case Routing

Use Case	When to Use
Client Engagement	Active client work, consulting engagements, fractional roles
Product Development	Building a product, course, app, or offering
Operational Domain	Internal workflows, business functions, recurring processes
Research & Learning	Exploration, skill-building, knowledge synthesis

Shared Architecture

All instructions follow the same structural pattern:

1. CONTEXT (~100-150 words)
   - Who/what this serves
   - Current state and goals
   - Key constraints

2. MANDATE (~50-100 words)
   - 4-5 bullet primary directives
   - What Claude must always do

3. MODES/WORKFLOWS (~100-200 words)
   - Operating modes or process stages
   - When to use each

4. GUARDRAILS (~75-125 words)
   - Always/Never rules
   - Scope boundaries
   - Escalation triggers

5. DOMAIN KNOWLEDGE (~100-200 words)
   - Key terms, entities, relationships
   - Tables preferred over prose

Target Length: 800-1,100 words total (fits comfortably in context)

Instructions vs Memory vs Files

Type	Purpose	Examples
Instructions	HOW to behave	Mandate, modes, guardrails, tone
Memory	WHAT is true now	Current status, recent decisions, active blockers
Project Files	Reference material	Templates, SOWs, client docs, pricing guides

Rule: Instructions define behavior. Memory holds state. Files provide reference.

Token Optimization

Apply these condensation patterns:

Tables over paragraphs — Structured data compresses 40-60%
Eliminate hedge words — "generally", "typically" → specific statements
Merge related bullets — Combine similar items into one
Reference, don't repeat — Point to skills/files for details
Symbols for status — 🔴 ⚠️ ✅ ⏳ instead of words
Implicit > explicit — Trust Claude's judgment on obvious matters

Output Format

When generating instructions:

Ask clarifying questions first if inputs are incomplete
Output as Markdown — Paste-ready for Claude Project instructions field
Include memory bootstrap — Suggested initial memory entries
Flag project files needed — What reference docs should be uploaded

Quality Checklist

Before finalizing instructions, verify:

Under 1,100 words total
Clear mandate (4-5 bullets max)
Modes/workflows defined with triggers
Guardrails include scope boundaries
Tables used where appropriate
No redundancy with default Claude behavior
Memory bootstrap provided
Project file recommendations included

Customization

See references/customize-this-skill.md to add your own templates and examples.

Download now (ZIP)

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

What did you use it for? e.g. Reviewing pull requests in a Rails app

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.

Check your setup, free