tp-run-full-design

tp-run-full-design is a skill for Claude Code from CurtisThe/three-pillars-plugin. It costs 42 tokens per session (29,010 once invoked), scanned A, original, Apache-2.0.

An automatic coordinator for a single task in the three-pillars TDD pipeline. TDD, or test-driven development, is a practice of using tests to guide implementation; here the coordinator moves through design, coding, validation, audits, and a pull request without waiting at each stage.

In plain words
What is it for?
Use it to pick up a task, run its design process, assign implementation phases to isolated workspaces, validate the results, run consolidation checks, close out the work, and prepare a completion pull request.
Why use it?
It connects the stages of one task and keeps decisions in a log, so developers can review the work afterward instead of manually coordinating every handoff.

Skill for Claude Code

Written for Claude Code: argument-hint in frontmatter. Also seen: mentions CLAUDE.md; mentions subagents; mentions Claude Code.

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

Good fit Use it to pick up a task, run its design process, assign implementation phases to isolated workspaces, validate the results, run consolidation checks, close out the work, and prepare a completion pull request.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/curtisthe/three-pillars-plugin/tp-run-full-design
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.

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

Made for: Claude Code.

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-run-full-design

README.md
[![agentmods](https://agentmods.dev/badge/skills/curtisthe/three-pillars-plugin/tp-run-full-design.svg)](https://agentmods.dev/skills/curtisthe/three-pillars-plugin/tp-run-full-design)
Your own site
<a href="https://agentmods.dev/skills/curtisthe/three-pillars-plugin/tp-run-full-design"><img src="https://agentmods.dev/badge/skills/curtisthe/three-pillars-plugin/tp-run-full-design.svg" alt="Measured on agentmods" height="20"></a>
Per session 42 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 29,010 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00042 $0.29010
Opus 5 $0.00021 $0.14505
Sonnet 5 $0.00008 $0.05802
Haiku 4.5 $0.00004 $0.02901

Measured 8d ago against content hash 2e65227afc65, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

tp-run-full-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 8d ago.

The scan reads SKILL.md. This mod also ships 23 executable files (scripts/build_synth_prompt.py, scripts/cleanup_worker_worktree.py, scripts/parse_candidate_response.py, …), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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-run-full-design/SKILL.md · 1,089 lines

How it starts

The opening of the file, as written. The whole thing — 1,089 lines — stays where its author put it; the contents beside it link to each section on GitHub.

tp-run-full-design — Autonomous Orchestrator (Mode A0, single candidate)

This skill drives the entire three-pillars TDD pipeline unattended for a single task. It dispatches each tier into a dedicated subagent (## Dispatch loop): every delegated --auto skill runs inline inside its slot's subagent, which synthesizes a clipped return envelope (## Per-slot return contract). The orchestrator itself only ever reads those envelopes plus the running token total — never the tier work product — which is what keeps it under ~100k regardless of design size. Every decision is logged to decisions.md per skills/_shared/auto-mode.md. (This supersedes the earlier inline prose-orchestration pattern — modelled on skills/tp-spike-auto/SKILL.md — where the orchestrator read each delegated SKILL.md and followed its --auto instructions in its own context.)

Mode A0 — MVP scope: Tier 1 (pickup) → Tier 2 (design pipeline) → Tier 3 (a worker Agent per plan phase, each in an isolated worktree) → Tier 3.5 (validation gate on the worker's structured response) → Tier 5 (consolidation audits) → Tier 5.6 (closeout: fold candidate→tp/{slug}, learn-verify, archive) → Tier 6 (completion PR tp/{slug}→{default}, design already closed) → Tier 7 (PR-iterate: drive the review loop to a reviewed-stable state). Tier 7 is on by default (--no-iterate opts out; --no-review implies no iteration) so a run ends at a reviewed-stable PR, not a fresh un-reviewed one — the orchestrator owns the review loop instead of leaving it for a separate invocation. The merge-only gate is preserved: Tier 7 iterates and pushes fix commits, but never merges — a human still merges the PR. Tier 4 (council over multiple candidates) is out of scope for MVP per design.md §Behavior 4 — adding it requires /council to gain a code-candidate evaluation mode, tracked as a follow-on design.

This orchestrator is not itself an --auto skill — it is the orchestrator of --auto skills. It does not expose --auto in its argument-hint, and the --autoauto-mode.md linkage requirement deliberately does not apply to this file (it is the orchestrator, not an auto-mode participant). It does, however, write [tp-run-full-design/tier-N]-prefixed entries to decisions.md per the prefix convention adopted in detailed-design.md §Decisions (OQ5).

Read the full file on GitHub · 1,089 lines

Files

What ships with it

34 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. 8d ago First seen · 1,089 lines · 42 tokens per session scan A 2e65227afc65

Subscribe to this mod's changes

tp-run-full-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 42 tokens to every session and 29,010 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.