sim-use: Skill for Claude Code

.agents/skills/write-ai-docs/SKILL.md

write-ai-docs is a skill for Claude Code, Codex from lycorp-jp/sim-use. It costs 115 tokens per session (1,896 once invoked), scanned A, original, Apache-2.0.

A documentation workflow for recording substantial software changes and investigations in numbered files under docs/ai/.

In plain words
What is it for?
Use it before and after cross-cutting changes, new platforms or transports, contract changes, investigations, and fixes whose root causes matter later.
Why use it?
It preserves the reasons behind technical decisions so later developers and agents do not have to reconstruct them from code or memory.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: installed under .agents/ (shared by several agents); mentions AGENTS.md.

This is lycorp-jp/sim-use's own configuration. It tells Claude Code and Codex how to work on sim-use itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything sim-use configures →

Needs its repository: it reads a path above its own folder, which exists only inside the repository. The line is | **Related** | issue #n, [NNN-other](../NNN-other/000-plan.md) |.

About the project

lycorp-jp/sim-use is a command-line tool that lets AI agents observe and control iOS simulators, Android emulators, and devices through accessibility information and input actions. It helps agents test mobile applications by reading compact screen descriptions and interacting with named elements, while the catalogue skills and instructions teach agents how to use those capabilities.

lycorp-jp/sim-use · 1,532 stars · on GitHub

Reuse

Borrowing it

Nothing to install: this file belongs to lycorp-jp/sim-use. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/lycorp-jp/sim-use/main/.agents/skills/write-ai-docs/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/lycorp-jp/sim-use

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 write-ai-docs

README.md
[![agentmods](https://agentmods.dev/badge/skills/lycorp-jp/sim-use/write-ai-docs/github.svg)](https://agentmods.dev/skills/lycorp-jp/sim-use/write-ai-docs)
Your own site
<a href="https://agentmods.dev/skills/lycorp-jp/sim-use/write-ai-docs"><img src="https://agentmods.dev/badge/skills/lycorp-jp/sim-use/write-ai-docs/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 write-ai-docs

Your own site · 80×15
<a href="https://agentmods.dev/skills/lycorp-jp/sim-use/write-ai-docs"><img src="https://agentmods.dev/badge/skills/lycorp-jp/sim-use/write-ai-docs.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 115 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,896 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 $0.00115 $0.01896
Opus 5.5 $0.00046 $0.00758
Sonnet 5.5 $0.00023 $0.00379
Haiku 4.5 $0.00012 $0.00190

Measured 16d ago against content hash db4f89de71e0, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

write-ai-docs 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 16d 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.

.agents/skills/write-ai-docs/SKILL.md · 177 lines

How it starts

The opening of the file, as written. The whole thing — 177 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Write AI docs

sim-use uses AI-docs-driven development. A substantial change starts with a record in docs/ai/, and each decision is written down when it is made, not reconstructed later. The record answers "why is it built this way?" for the next human or agent. Read docs/ai/README.md first: it has the index and the folder layout.

When to write

A new entry (docs/ai/NNN-<slug>/) is necessary when the work:

  • adds a verb, a platform, a transport, or a backend, or changes how a verb reaches a device (dispatch, daemon, HID transport, bridge);
  • changes a contract: the daemon wire protocol, the bridge wire protocol (PROTOCOL_VERSION), the --json envelope, the bundled skill's compatibility floor (MINIMUM_SIM_USE_VERSION), or the host-environment matrix;
  • starts as an investigation (a new Xcode, a reproducible failure, an upstream idb change) and ends in a decision;
  • fixes a bug whose root cause or fix shape must guide later code (for example, "HID from a short-lived process is lost on the dtuhidd transport").

An amendment to an existing entry is necessary when follow-up work, a review round, or a later finding changes what the entry says. Look for an existing entry on the topic before you open a new one.

No record is necessary for typo or formatting fixes, dependency bumps, test-only or fixture-only changes, small fixes that need no separate design record, reviews, and status reports. Keep changelog entries focused on user impact, not root-cause detail. If you are not sure, ask: will someone later need the reasoning to change this code safely? If yes, write it down.

Workflow

1. New entry: plan first

  1. Get the next number: ls docs/ai | sort, then take the highest NNN + 1 (three digits). Numbers are never reused.
  2. Before you write code, create docs/ai/NNN-<kebab-slug>/000-plan.md from the template below, with Status: Planned. Use 000-investigation.md when the work starts as a root-cause hunt: record the reproduction, the evidence and the hypotheses, then the decision.
  3. Add the entry to the index table in docs/ai/README.md.
  4. Implement the work in the normal branch and PR flow. When you make a decision that the plan did not foresee, record it at once in the plan's Decisions table, or in an amendment.
  5. After the work lands, write 001-action.md: what shipped, the PRs, the key files, and the deviations from the plan. Set the plan status to Implemented. Ship the docs in the same PR as the code when you can.

Read the full file on GitHub · 177 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. 16d ago Changed db4f89de71e0
  2. 23d ago First seen · 177 lines · 115 tokens per session scan A ffd7bb37d041

Subscribe to this mod's changes

write-ai-docs is a skill published in the GitHub repository lycorp-jp/sim-use (1,532 stars, last pushed 13d ago), licensed Apache-2.0. It adds 115 tokens to every session and 1,896 once invoked, about $0.0005 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 2026-10-15.

Related

write-ai-docs compared with skills from other repositories

write-ai-docs next to 6 skills from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
write-ai-docslycorp-jp/sim-useSkill1.5k13d agoApache-2.0A115Claude Code, Codex
open-gui-remote-controlCore-Mate/OpenGUISkill1.8k19d agononeA66Codex
open-gui-bootstrapCore-Mate/OpenGUISkill1.8k19d agononeA60Codex
opengui-coremate-installCore-Mate/OpenGUISkill1.8k19d agononeA48Codex
opengui-plugin-installCore-Mate/OpenGUISkill1.8k19d agononeA43Claude Code, Codex
controlCore-Mate/OpenGUISkill1.8k19d agononeA55Codex
web-designmartineserios/thebranaSkill3todayMITA24Claude Code