devtest

devtest is a skill for Claude Code, Codex from AILiteracyLab/Claude-Build-Test-Loop. It costs 58 tokens per session (9,954 once invoked), scanned A, original, MIT.

A three-agent loop for building and checking digital features against a specification. One agent builds, another tests, and a supervisor watches for stalled work.

In plain words
What is it for?
Use it for user interfaces, APIs, command-line tools, conversational AI, data pipelines, and other digital functions that need implementation and validation.
Why use it?
It keeps implementation tied to the agreed requirements and catches mismatches during repeated review. It also notices when agents freeze or stop handing work over.

Skill for Claude CodeCodex

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/ailiteracylab/claude-build-test-loop/skill
Any agent
npx skills add AILiteracyLab/Claude-Build-Test-Loop --skill skill
Clone the repo
git clone --depth 1 https://github.com/AILiteracyLab/Claude-Build-Test-Loop

Made for: Claude Code, Codex.

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 devtest

README.md
[![agentmods](https://agentmods.dev/badge/skills/ailiteracylab/claude-build-test-loop/skill.svg)](https://agentmods.dev/skills/ailiteracylab/claude-build-test-loop/skill)
Your own site
<a href="https://agentmods.dev/skills/ailiteracylab/claude-build-test-loop/skill"><img src="https://agentmods.dev/badge/skills/ailiteracylab/claude-build-test-loop/skill.svg" alt="Measured on agentmods" height="20"></a>
Per session 58 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 9,954 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. 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 $0.00058 $0.09954
Opus 5 $0.00029 $0.04977
Sonnet 5 $0.00012 $0.01991
Haiku 4.5 $0.00006 $0.00995

Measured 3d ago against content hash 97001d06a698, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

devtest scanned grade A with 1 finding 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 3d 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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

| **API/Backend** | Expected request/response pairs, status codes, data shapes | HTTP requests (curl, fetch, Playwright API calls), response validation |
skill/SKILL.md · 466 lines

How it starts

The opening of the file, as written. The whole thing — 466 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Design-Test Skill

A general-purpose three-agent feedback loop. The user (Designer) sets the spec, one agent (Builder) implements it, another agent (Tester) validates the result against the spec, and a third agent (Supervisor) monitors for frozen agents and stalled handoffs. They iterate until the Tester confirms the implementation matches. This process applies to any digital function — not just visual design.

Roles

Role Who Responsibility
Designer The human user Defines the spec — what "correct" looks like. Has final say.
Builder Agent teammate Implements and refines the code/system to match the Designer's spec.
Tester Agent teammate Validates the result against the spec using appropriate testing methods. Reports discrepancies or approves.
Supervisor Agent teammate Monitors Builder and Tester every 3 minutes. Detects frozen agents, stalled handoffs, and tasks that should have been marked done. Nudges agents to continue or escalates to the team lead.

Workflow

Designer provides spec
        |
        v
   Builder implements (Round 1)
        |
        v
   Tester validates against spec (Round 1)
        |
    Match? --PASS--> Team lead validates evidence --> Done! Report to Designer
        |
   CONDITIONAL PASS? --> Team lead presents minor issues to Designer
        |                     |
        |               Accept? --YES--> Done!
        |                     |
        |                     NO (continue iterating)
        |                     |
        FAIL  <---------------+
        |
        v
   Tester sends feedback to Builder
        |
        v
   Builder checkpoints, revises (Round N), notifies Tester
        |
        v
   (loop back to Tester - Round N)

   [Supervisor monitors every 3 min throughout]
   [Builder & Tester check Supervisor liveness every 3 rounds]

What Can Be Spec'd and Tested

This skill is not limited to visual design. The Designer can provide any kind of spec:

Read the full file on GitHub · 466 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. 3d ago First seen · 466 lines · 58 tokens per session scan A 97001d06a698

Subscribe to this mod's changes

devtest is a skill published in the GitHub repository AILiteracyLab/Claude-Build-Test-Loop (4 stars, last pushed 6mo ago), licensed MIT. It adds 58 tokens to every session and 9,954 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other skills, from other repositories

relay

Start, monitor, steer, or stop an autonomous multi-session build run. Use for "/relay init ", "/relay run", "/relay status", "/relay stop", "/relay note", "/relay resume", "/relay doctor", or any request to run a long multi-phase plan to completion unattended.

cptnbg/relay · 63 tokens

fleet-manager

Use when working with remote Linux servers over SSH through the fleet-manager commands — onboarding or switching servers, deploying SSH keys, checking connectivity, disk, memory or journal logs, transferring files, or managing Docker Compose stacks across a fleet of hosts.

CallMeTechie/fleet-manager · 51 tokens

google-ads-audit

Google Ads account audit and business context setup. Run this first — it gathers business information, analyzes account health, and saves context that all other ads skills reuse. Trigger on "audit my ads", "ads audit", "set up my ads", "onboard", "account overview", "how's my account", "ads health check", "what should…

nowork-studio/notfair-plugin · 114 tokens

review

5-pass structured code review — correctness, security, performance, readability, consistency.

SethGammon/Citadel · 17 tokens

alive:system-upgrade

Upgrade ALIVE to the current version. Handles v1/v2/v3.x source states, multi-surface aware (alive-mcp / Hermes / Codex), retroactive version detection, partial-failure resume, dry-run previews, and rollback inspection.

alivecontext/alive · 57 tokens

codex-test-gen

Generate unit tests for specified functions using Codex MCP.

sd0xdev/sd0x-harness · 15 tokens