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.
npx skills add jpavley/meta-loop-ios --skill meta-loop-methodologygit clone --depth 1 https://github.com/jpavley/meta-loop-iosWrote 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.
[](https://agentmods.dev/skills/jpavley/meta-loop-ios/meta-loop-methodology)<a href="https://agentmods.dev/skills/jpavley/meta-loop-ios/meta-loop-methodology"><img src="https://agentmods.dev/badge/skills/jpavley/meta-loop-ios/meta-loop-methodology/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/jpavley/meta-loop-ios/meta-loop-methodology"><img src="https://agentmods.dev/badge/skills/jpavley/meta-loop-ios/meta-loop-methodology.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00021 | $0.04992 |
| Opus 5.5 | $0.00008 | $0.01997 |
| Sonnet 5.5 | $0.00004 | $0.00998 |
| Haiku 4.5 | $0.00002 | $0.00499 |
Grade A, and why
meta-loop-methodology 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 1mo 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.
How it starts
The opening of the file, as written. The whole thing — 387 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Software Development Process with Agentic Interfaces
Summary: This document describes a "meta loop" where a single experienced developer collaborates with AI coding agents (like Claude Code) to achieve single-shot repeatability and transferability—the ability to deliver production-ready software in one attempt, reproducibly, across different platforms. The process flows through three loops (Wireframe → Specification → Implementation), each using four modes: Human Action (manual work), Plan Mode (AI creates structured plans), Ask Mode (AI reviews and provides feedback), and Agent Mode (AI executes the plan). The key insight is "delete and regenerate"—when errors occur, artifacts are deleted rather than patched, and the documentation is improved so the AI can regenerate correctly. This inverts traditional economics: code becomes cheap to regenerate while documentation becomes the persistent value that preserves learning across sessions.
Introduction
The goal of this software development process (meta loop) is for a single developer (the user) collaborating with one or more AI coding agents to deliver high-quality production-ready software applications and systems efficiently.
In this context "AI coding agent" means an agentic interface to a family of frontier models trained, fine-tuned, system prompted, with interfaces to software development capabilities. An "agentic interface" has action capabilities, observation capabilities, variable autonomy, and a "loop" ( a "while loop" with tool execution and context accumulation repeats until there is a reason to stop). The success and failure of AI coding is tied up in helping the agent understand what the (1) tasks is, (2) how to accomplish it, and (3) the "definition of done". Ironically 1, 2, and 3 are problems that have plagued software development from the beginning. There are no good answers as every software development team's understanding of 1, 2, and 3 are subjective and changing and not objective and fixed. Even with specs, plans, and tests, 1, 2, and 3 are fuzzy. This is because a map is not the territory, no plan survives meeting the enemy, and "done", like beauty, is in the eye of the beholder.
What ships with it
1 file 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.
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.
- 1mo ago First seen · 387 lines · 21 tokens per session scan A 4ca60a1075e5
meta-loop-methodology is a skill published in the GitHub repository jpavley/meta-loop-ios (3 stars, last pushed 8mo ago), licensed MIT. It adds 21 tokens to every session and 4,992 once invoked, about $0.0001 per session on Opus 5.5. 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.
Other skills, from other repositories
ax-media
Now Playing, AVFoundation audio, haptics, photo library, MusicKit, CarPlay, and spatial audio patterns for iOS.
ax-performance
Use when app feels slow, memory grows, hangs occur, frame rate issues, or profiling needed - covers Instruments workflows, memory debugging, hang diagnostics, display performance, MetricKit, xctrace CLI, and ObjC block retain cycles.
ax-3d-games
2D/3D game and spatial content development -- SpriteKit (2D games), SceneKit (3D legacy, deprecated iOS 26), RealityKit (modern 3D/AR/ECS), physics, ECS architecture, SpriteView/RealityView SwiftUI integration, and diagnostics.
ax-privacy
Privacy manifests, permissions UX, ATT, localization (String Catalogs, plurals, RTL, Xcode 26 generated symbols), Transferable/drag-drop/copy-paste/ShareLink, App Intents discoverability, debug deep links, and Required Reason APIs.
ax-background-tasks
BGTaskScheduler, background URL sessions, silent push, beginBackgroundTask, and BGContinuedProcessingTask patterns for iOS.
ax-camera
Camera capture with AVFoundation -- AVCaptureSession, SwiftUI preview, RotationCoordinator (iOS 17+), responsive capture (zero-shutter-lag, deferred processing), session interruptions, camera switching, video recording, and diagnostics.