agentmods.dev

ulw-loop

ulw-loop is a skill for Codex from daeryundf2-prog/LAZYANTIGRAVITY. It costs 24 tokens per session (1,819 once invoked), scanned B, original, MIT.

A repeating work process that breaks a goal into smaller steps and checks each step against evidence. Ultrawork mode is the operating mode it uses for this process.

In plain words
What is it for?
It helps decompose goals, investigate their parts systematically, and use evidence to guide the work.
Why use it?
It provides a structured way to handle larger tasks and reduce unsupported conclusions.

Skill for Codex

Written for Codex: agents/openai.yaml present. Also seen: mentions subagents.

not rated 14repo changed 14d ago B scan 24 tokens original MIT

Good fit It helps decompose goals, investigate their parts systematically, and use evidence to guide the work.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/daeryundf2-prog/lazyantigravity/ulw-loop
View source ↗ daeryundf2-prog/LAZYANTIGRAVITY
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 daeryundf2-prog/LAZYANTIGRAVITY --skill ulw-loop
Clone the repo
git clone --depth 1 https://github.com/daeryundf2-prog/LAZYANTIGRAVITY

Made for: 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 ulw-loop

README.md
[![agentmods](https://agentmods.dev/badge/skills/daeryundf2-prog/lazyantigravity/ulw-loop/github.svg)](https://agentmods.dev/skills/daeryundf2-prog/lazyantigravity/ulw-loop)
Your own site
<a href="https://agentmods.dev/skills/daeryundf2-prog/lazyantigravity/ulw-loop"><img src="https://agentmods.dev/badge/skills/daeryundf2-prog/lazyantigravity/ulw-loop/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.

agentmods 80×15 button for ulw-loop

Your own site · 80×15
<a href="https://agentmods.dev/skills/daeryundf2-prog/lazyantigravity/ulw-loop"><img src="https://agentmods.dev/badge/skills/daeryundf2-prog/lazyantigravity/ulw-loop.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 24 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,819 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 1 finding. 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 $0.00024 $0.01819
Opus 5.5 $0.00010 $0.00728
Sonnet 5.5 $0.00005 $0.00364
Haiku 4.5 $0.00002 $0.00182

Measured 14d ago against content hash eeb31ee5e631, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade B, and why

ulw-loop scanned grade B 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 14d 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.

Reads agent configuration directoriesmediumAgent snooping

.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.

3. Resolve the ULW CLI via Bootstrap (PLUGIN_ROOT / Windows PowerShell / `~/.gemini/config/plugins/lazyantigravity/.../cli.js`). If CLI is missing, stop and report — do not hand-edit goal JSON.
components/ulw-loop/skills/ulw-loop/SKILL.md · 111 lines

How it starts

The opening of the file, as written. The whole thing — 111 lines — stays where its author put it; the contents beside it link to each section on GitHub.

ulw-loop

Use this skill when the user asks for ulw-loop, ulw, durable goal execution, evidence-led work, manual QA, or checkpointed long-running delivery.

This skill is intentionally compact. The full workflow lives in references/full-workflow.md. Read only the sections needed for the current phase, then execute them exactly.

Default host for LazyAntigravity: Google Antigravity + Gemini 3.8 Flash (High).

Required First Steps

  1. Open references/full-workflow.md.
  2. Read Runtime selection, Bootstrap, Execution Loop, and Manual-QA channels before running any ULW command or recording evidence.
  3. Resolve the ULW CLI via Bootstrap (PLUGIN_ROOT / Windows PowerShell / ~/.gemini/config/plugins/lazyantigravity/.../cli.js). If CLI is missing, stop and report — do not hand-edit goal JSON.
  4. If the task has code edits, tests, QA, or commit work, follow the full workflow's delegation and evidence rules. Tests alone never prove done.
  5. If the work mutates the workspace, take a baseline snapshot BEFORE edits: prefer the session_tree_snapshot MCP tool (workspace server) when available, otherwise run node "$PLUGIN_ROOT/components/session-tree/dist/cli.js" snapshot "<label>". Verify .lazyantigravity/session-tree/nodes.json gained a node — a JSON note written anywhere else is NOT a snapshot and satisfies no criterion.

Non-Negotiables

  • Use the ulw-loop CLI state under .omo/ulw-loop; do not hand-edit goal state.
  • After any compaction or context loss, re-read brief + goals + ledger FIRST (read files directly) plus lazyantigravity ulw-loop status --json, then resume; never re-plan from scratch.
  • Every success criterion needs observable evidence from a real channel: tmux, HTTP, browser, or computer-use.
  • Record evidence through the CLI only after cleanup receipts are available.
  • Delegate via invoke_subagent only (see ../references/antigravity-tools.md).
  • When invoking a subagent, pass a role envelope:
    • mayFinalizeRun=false
    • mayModifyGlobalRunState=false
    • mustReturn=SubagentResultEnvelope
    • requiresParentAck=true
    • Do not claim the whole /ulw task is complete.
    • Do not mark run as completed or failed.
  • Pass Subagents[].Model on invoke_subagent (canTierRoute=true, hostEnforced=false):
    • plan / research / implement / explore → Model: "flash"
    • verify / adversarial review → Model: "inherit" (when session is Claude 5.5, child stays on Claude) / Model: "pro" (when session is Gemini)
    • tiny repetitive chores → Model: "flash_lite"
    • inherit parent → Model: "inherit"
  • Follow the user-selected session model (supported: Gemini 3.8 Flash (High), Claude Opus 5.5, Claude Sonnet 5.5). Prefer tier routing over switching the whole session UI. Manual UI switch to Gemini 3.1 Pro or Claude Opus 5.5 is optional when you want the parent itself on Pro/Opus.
  • Use git-master for git-tracked edits: inspect recent and touched-path commit history, then commit each verified work unit atomically.

Read the full file on GitHub · 111 lines

Files

What ships with it

8 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. 14d ago Changed · +2 lines eeb31ee5e631
  2. 24d ago Changed · +4 lines db776ea6333d
  3. 1mo ago Changed 208f8dfa11a8
  4. 1mo ago First seen · 105 lines · 24 tokens per session scan B f59931a17e3c

Subscribe to this mod's changes

ulw-loop is a skill published in the GitHub repository daeryundf2-prog/LAZYANTIGRAVITY (14 stars, last pushed 7d ago), licensed MIT. It adds 24 tokens to every session and 1,819 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

PlanningTypeScriptagentic-codingai-agentautomationdeveloper-toolslazycodex
Related

Other skills, from other repositories

top-down-typescript

TypeScript and JavaScript coding style conventions, centered on top-down narrative ordering (caller first, helpers below) and functions over classes. Read before writing or reviewing TypeScript/JavaScript code, including code inside a spec or a plan.

paleo/alignfirst · 53 tokens

loom-typescript

TypeScript language expertise for type-safe, production-quality code.

cosmix/loom · 16 tokens

ai-provider-anthropic-sdk

Official Anthropic SDK patterns for TypeScript/Node.js — client setup, Messages API, streaming, tool use, vision, extended thinking, structured outputs, prompt caching, batch API, and production best practices.

agents-inc/skills · 48 tokens

ai-infrastructure-ollama

Local LLM inference with the Ollama JavaScript client -- chat, streaming, tool calling, vision, embeddings, structured output, model management, and OpenAI-compatible endpoint.

agents-inc/skills · 41 tokens

ai-infrastructure-replicate

Replicate SDK patterns for TypeScript/Node.js -- client setup, predictions, streaming, webhooks, file handling, model versioning, deployments, and training.

agents-inc/skills · 39 tokens

ai-infrastructure-together-ai

Together AI SDK patterns for TypeScript — client setup, chat completions, streaming, structured output, function calling, embeddings, image generation, fine-tuning, and OpenAI-compatible endpoints.

agents-inc/skills · 44 tokens