tp-design

tp-design is a skill for Claude Code, Codex from CurtisThe/three-pillars-plugin. It costs 36 tokens per session (2,598 once invoked), scanned A, original, Apache-2.0.

A conversation-based planning tool for creating a high-level design document for a test-driven development project. It saves the document as design.md in a named project folder.

In plain words
What is it for?
Use it to discuss, create, or revise the high-level design for a project whose development is guided by tests.
Why use it?
It turns an initial idea into a written design before implementation begins, so the project has a shared plan to work from.

Skill for Claude CodeCodex

Part of the three-pillars plugin — 37 skills, 20 agents shipped together

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/curtisthe/three-pillars-plugin/tp-design
Any agent
npx skills add CurtisThe/three-pillars-plugin --skill tp-design
Clone the repo
git clone --depth 1 https://github.com/CurtisThe/three-pillars-plugin

Made for: Claude Code, Codex.

Or install three-pillars, the plugin that ships this one along with the rest of its 37 skills, 20 agents.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for tp-design

README.md
[![agentmods](https://agentmods.dev/badge/skills/curtisthe/three-pillars-plugin/tp-design.svg)](https://agentmods.dev/skills/curtisthe/three-pillars-plugin/tp-design)
Your own site
<a href="https://agentmods.dev/skills/curtisthe/three-pillars-plugin/tp-design"><img src="https://agentmods.dev/badge/skills/curtisthe/three-pillars-plugin/tp-design.svg" alt="Measured on agentmods" height="20"></a>
Per session 36 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,598 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00036 $0.02598
Opus 5 $0.00018 $0.01299
Sonnet 5 $0.00007 $0.00520
Haiku 4.5 $0.00004 $0.00260

Measured 4d ago against content hash d15ce05efc9e, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

tp-design scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/tp-design/SKILL.md · 109 lines

How it starts

The opening of the file, as written. The whole thing — 109 lines — stays where its author put it; the contents beside it link to each section on GitHub.

High-Level Design

Create or revise the high-level design for a TDD project through conversation with the user.

Argument: {design-name} (required) — kebab-case name, becomes the directory under three-pillars-docs/tp-designs/.

Steps

  1. Run first-run preflight per skills/_shared/first-run.md.

  2. Resolve the design directory: three-pillars-docs/tp-designs/{design-name}/. Create it if it doesn't exist.

  3. Run collaboration preflight per skills/_shared/collaboration.md with phase: "design". This verifies the branch and acquires or refreshes the lock for this design. Honor --force-takeover if passed.

    Seat-context note: if the collaboration preflight notes that you are on the base branch ({base} / master), treat that as an affirmation — being on {base} in the resolved seat (the base checkout / worktree host) is the correct coordination point for spinning up a new design worktree. It is not a reason to pause or seek an alternative checkout. See skills/_shared/topology.md for the canonical seat and worktree layout; the collaboration preflight's branch check is seat-aware — in the seat it offers worktree provisioning (provision-and-instruct), never an in-place checkout.

2b. Update .claude/last-design MRU — now that the lock/branch are claimed, run the MRU snippet at the bottom of skills/_shared/validate-name.md (the bash that prepends the design name, dedupes, caps at 10). This is the moment a subsequent /clear + /tp-session-restore (no argument) needs to resolve to this design, not whatever was active before. The snippet handles the .gitignore append; do not git add the file. 2c. Repo-map preamble (optional) per skills/_shared/repo-map-preamble.md. If aider is on PATH, generate a structural map of the codebase to inform the design conversation; if absent, skip silently and proceed. 3. Read project context per skills/_shared/read-project-docs.md. Read three-pillars-docs/vision.md first — every question you ask in the design conversation should be framed against the vision's Problem, Users, Principles, and Non-goals. If three-pillars-docs/vision.md is missing, tell the user and recommend /tp-setup but don't block. If the roadmap has a ## Current Focus table, note where this new design fits relative to current priorities — mention this during the design conversation so the user can decide its priority. 4. Check for existing design.md. If it exists, read it and ask the user whether they want to revise it or start fresh. If starting fresh, warn that downstream artifacts (detailed-design.md, plan.md) will become stale. 5. Vision alignment check. Before the main design conversation, explicitly ask: "How does this design advance the problem or principles stated in three-pillars-docs/vision.md?" Write the user's answer as the seed for the Problem section of design.md. If the answer is weak or the design obviously touches a stated non-goal, surface that tension now — it is much cheaper to reject or reshape a design at this stage than to fight it through detailed-design and audit later. If three-pillars-docs/vision.md doesn't exist, skip this step but note it. 5b. Declare the weight class. Read weight-class from the frontmatter of three-pillars-docs/tp-designs/{design-name}/seed.md if the seed exists and carries one. If absent, ask the user once, rubric-assisted: score the four axes (risk, blast radius, reversibility, novelty — each low/medium/high) and run python3 "$TP_ROOT"/skills/_shared/weight_class.py recommend --risk … --blast-radius … --reversibility … --novelty … for the recommendation, then let the user confirm or override. Protocol reference: skills/_shared/weight-class.md. The declared class is stamped onto design.md in step 7 and propagates to every artifact generated downstream.

Read the full file on GitHub · 109 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 4d ago First seen · 109 lines · 36 tokens per session scan A d15ce05efc9e

Subscribe to this mod's changes

tp-design is a skill published in the GitHub repository CurtisThe/three-pillars-plugin (4 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 36 tokens to every session and 2,598 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.