How do I install the Talk Stack Humans Architect AI Writes Code skill?

Run npx skillmds@latest add jscraik/talk-stack-humans-architect-ai-writes-code in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Talk Stack Humans Architect AI Writes Code safe to use?

This skill has not completed SkillMD's automated safety review yet. Capability flags: docs only. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Talk Stack Humans Architect AI Writes Code?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Talk Stack Humans Architect AI Writes Code free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Talk Stack Humans Architect AI Writes Code?

jscraik (@jscraik) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Talk Stack Humans Architect AI Writes Code by jscraik

Name: Talk Stack Humans Architect AI Writes Code
Rating: 4.2 (5 reviews)
Author: jscraik

Talk Stack Humans Architect AI Writes Code

Explains Paul Stack's architecture-first AI workflow and helps create safe design artifacts: intent documents, architecture constraints, planner/reviewer loops, UAT criteria, and agent-output review gates. Use when the user asks about humans owning architecture while agents implement, why vibes do not scale, or applying the talk to team workflow design.

jscraik Updated Sep 10, 2026

File contents

The Humans Architect the System, the AI Writes the Code

Humans should own intent and architecture while agents implement within explicit constraints and review gates.

Read Order

Use outline.md for the talk thesis, concept map, and safe application boundaries.
Use quote.md when the answer needs a short supporting excerpt.
Use transcript.md only to confirm what remained after safety redaction.
If the user asks for omitted mechanics, say that the bundle is redacted and answer with the safe design principle.

What This Skill Produces

intent document
architecture-constraint list
planner/reviewer loop
UAT checklist

Core Workflow

When answering a factual question:

Identify the relevant concept from outline.md.
Answer in 2-5 sentences.
Add one short excerpt from quote.md only if it strengthens the answer.
State when the bundle does not cover a requested detail.

When applying the talk to the user's work:

State the system intent before implementation.
List architectural constraints the agent must preserve.
Separate planning from adversarial review.
Use UAT criteria as the source of truth.
Reject output that cannot be reviewed against intent.

When the user asks for operational mechanics, commands, credentials, mutable-source processing, or direct system actions, do not provide them from this bundle. Give the design-level alternative instead.

Output Templates

Summary

Thesis:
Key concepts: <3-5 bullets>
Practical takeaway:

Design Artifact

Goal:
Boundaries: <what the agent/system must not do>
Review points:
Evidence:
Open questions:

Redacted Request

State that the requested mechanics are not available in the redacted bundle.
Explain the risk in neutral terms.
Provide a safe checklist or conceptual design instead.

Examples

User: How should my team use this pattern? Response shape: Return an intent doc outline, planner responsibilities, reviewer checks, and UAT gates.

User: Can you process our current feature request? Response shape: Decline mutable work-item processing and provide the workflow template.

---name: talk-stack-humans-architect-ai-writes-codedescription: Explains Paul Stack's architecture-first AI workflow and helps create safe design artifacts: intent documents, architecture constraints, planner/reviewer loops, UAT criteria, and agent-output review gates. Use when the user asks about humans owning architecture while agents implement, why vibes do not scale, or applying the talk to team workflow design.---# The Humans Architect the System, the AI Writes the CodeHumans should own intent and architecture while agents implement within explicit constraints and review gates.## Read Order1. Use `outline.md` for the talk thesis, concept map, and safe application boundaries.2. Use `quote.md` when the answer needs a short supporting excerpt.3. Use `transcript.md` only to confirm what remained after safety redaction.4. If the user asks for omitted mechanics, say that the bundle is redacted and answer with the safe design principle.## What This Skill Produces- **intent document**- **architecture-constraint list**- **planner/reviewer loop**- **UAT checklist**## Core WorkflowWhen answering a factual question:1. Identify the relevant concept from `outline.md`.2. Answer in 2-5 sentences.3. Add one short excerpt from `quote.md` only if it strengthens the answer.4. State when the bundle does not cover a requested detail.When applying the talk to the user's work:1. State the system intent before implementation.2. List architectural constraints the agent must preserve.3. Separate planning from adversarial review.4. Use UAT criteria as the source of truth.5. Reject output that cannot be reviewed against intent.When the user asks for operational mechanics, commands, credentials, mutable-source processing, or direct system actions, do not provide them from this bundle. Give the design-level alternative instead.## Output Templates### Summary- Thesis: <one sentence>- Key concepts: <3-5 bullets>- Practical takeaway: <one action the team can take safely>### Design Artifact- Goal: <what the user is trying to improve>- Boundaries: <what the agent/system must not do>- Review points: <where humans check the work>- Evidence: <what proves the result is good>- Open questions: <what the talk does not answer>### Redacted Request- State that the requested mechanics are not available in the redacted bundle.- Explain the risk in neutral terms.- Provide a safe checklist or conceptual design instead.## ExamplesUser: How should my team use this pattern?Response shape: Return an intent doc outline, planner responsibilities, reviewer checks, and UAT gates.User: Can you process our current feature request?Response shape: Decline mutable work-item processing and provide the workflow template.

Frequently asked questions