think-first SKILL.md skillA

think-first from szoloth/skill-pack is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned
mdr add szoloth/skill-pack/think-first@git:20260303.cf416f2

Pins git:20260303.cf416f2 in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 17 of 17 checks passed
Kind
SKILL.md skill
Repository
szoloth/skill-pack
Path
skills/think-first/SKILL.md
Versions recorded
1
Latest version
git:20260303.cf416f2, committed Mar 3, 2026
Size
3,529 bytes
Repository stars
14
License
MIT

Cognitive engagement coach based on "Think First, AI Second" principles. This skill should be used when the user asks strategic, architectural, or high-stakes questions, OR when they explicitly request challenge/critique (e.g., "poke holes", "devil's advocate", "challenge this"). Promotes active thinking over passive AI consumption.

Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.

What the file says

# Think First, AI Second

*Based on [Ines Lee's article](https://every.to/p/think-first-ai-second) on maintaining cognitive engagement when using AI.*

## Core Principle

The order of operations matters: **think first, then use AI**. Research shows that starting with AI creates passive consumption patterns, while forming your own thoughts first maintains full cognitive activation AND gets efficiency benefits.

## Behavior Guidelines

### 1. For Big Decisions: Prompt for Initial Thinking

When the user asks strategic, architectural, or high-stakes questions, ask for their initial thinking before responding.

**Trigger phrases:**
- "How should I approach..."
- "What's the best way to..."
- "Should I..." (for significant decisions)
- Career, system architecture, major refactors

**Response pattern:**
> "What's your initial thinking on this?"

Or more specifically:
> "Before I weigh in—what's your instinct here? What options are you considering?"

**Skip this for:**
- Quick tactical questions
- Debugging
- "What does X mean"
- Factual lookups
- When user has already shared their thinking

### 2. Active Hole-Finding

Actively scan every request for:
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add szoloth/skill-pack/think-first@sha256:fdd936896f27ff71

The line at the top of this page pins the label git:20260303.cf416f2, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_vrpcvvewm2p4qdj5.svg)](https://markdownregistry.com/a/art_vrpcvvewm2p4qdj5)

3 badge image fetches in 30 days. What the badge shows.

Versions

The registry has recorded 1 version of this file, committed on Mar 3, 2026.

versioncommittedcommitsizeaudit
git:20260303.cf416f2 latest2026-03-03 cf416f2 3,529 BA view

Audit of the latest version

A  17 of 17 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.
  • pass: Frontmatter block present
  • pass: Frontmatter declares a name
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (3529 bytes)
  • pass: No zero-width or bidi control characters
  • pass: No instruction hidden inside an HTML comment
  • pass: No link to an exfiltration or paste host
  • pass: No credential-shaped string
  • pass: No instruction to send local credentials anywhere
  • pass: No text hidden with inline styles
  • pass: No prompt-injection phrasing
  • pass: No curl or wget piped into a shell
  • pass: No recursive delete of root, home or parent
  • pass: No instruction to read or print local credentials
  • pass: No base64 blob over 200 characters
  • pass: No link to a raw IP address
  • pass: No script tag

Source

GitHub

szoloth/skill-pack · 14 stars · license MIT · pushed 2026-09-09 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_vrpcvvewm2p4qdj5
GET https://markdownregistry.com/api/v1/resolve?ref=szoloth/skill-pack/think-first
GET https://markdownregistry.com/api/v1/blob/fdd936896f27ff7122b7be6fdab5bada0d4bf514d6965e1907f0d80821c59a85

Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.

Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.

More from szoloth/skill-pack

pstack-full SKILL.md skill
szoloth/skill-pack · .agents/skills/pstack-full/SKILL.md · Full Poteto Mode and project notes for nontrivial engineering in skill-pack. Apply when planning, building, debugging…
git:20260909.7427c15 · audit A · 14 stars
pstack-project SKILL.md skill
szoloth/skill-pack · .agents/skills/pstack-project/SKILL.md · Apply pstack to Skill pack maintenance using its actual controls and local evidence boundaries.
git:20260909.7427c15 · audit A · 14 stars
agentic-review SKILL.md skill
szoloth/skill-pack · skills/agentic-review/SKILL.md · Deep multi-agent code review for local changes. Inspired by AmpCode's agentic review. Use when you want comprehensive…
git:20260303.cf416f2 · audit A · 14 stars
b2b-expert-advisor SKILL.md skill
szoloth/skill-pack · skills/b2b-expert-advisor/SKILL.md · B2B startup strategy advisor grounded in Lenny Rachitsky's 7-part series and April Dunford's positioning framework…
git:20260313.1163d85 · audit A · 14 stars
design-swarm SKILL.md skill
szoloth/skill-pack · skills/design-swarm/SKILL.md · Orchestrate a team of 10 design agents for holistic UI audit, ideation, and implementation. Use when: "design swarm"…
git:20260303.302b93b · audit A · 14 stars
doc-coauthoring SKILL.md skill
szoloth/skill-pack · skills/doc-coauthoring/SKILL.md · Guide users through a structured workflow for co-authoring documentation. Use when user wants to write documentation…
git:20260327.c9b0fe1 · audit A · 14 stars
frontend-design SKILL.md skill
szoloth/skill-pack · skills/frontend-design/SKILL.md · Create distinctive, production-grade frontend interfaces with high design quality. Use this skill when the user asks to…
git:20260327.c9b0fe1 · audit A · 14 stars
iterate SKILL.md skill
szoloth/skill-pack · skills/iterate/SKILL.md · Diverge-converge workflow: spin up N agents on isolated worktrees, each exploring a different approach to the same…
git:20260303.cf416f2 · audit A · 14 stars
master SKILL.md skill
szoloth/skill-pack · skills/master/SKILL.md · Turn reference docs into active mastery through retrieval practice, case binding, and scenario simulation. Combines…
git:20260327.02d8df1 · audit A · 14 stars
mcp-builder SKILL.md skill
szoloth/skill-pack · skills/mcp-builder/SKILL.md · Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external…
git:20260327.c9b0fe1 · audit A · 14 stars
napkin SKILL.md skill
szoloth/skill-pack · skills/napkin/SKILL.md · Maintain a per-repo napkin as a continuously curated runbook (not a session log). Activates EVERY session. Read and…
v6.0.0 · audit A · 14 stars
orchestrating-swarms SKILL.md skill
szoloth/skill-pack · skills/orchestrating-swarms/SKILL.md · Master multi-agent orchestration using Claude Code's TeammateTool and Task system. Use when coordinating multiple…
git:20260327.c9b0fe1 · audit A · 14 stars

Every file in szoloth/skill-pack

Browse by kind, by grade A, or by owner.