texting-setup

A one-time setup skill for reading and sending iMessage or SMS messages from Claude on a Mac. It installs the imsg engine and guides the user through the required macOS permissions.

In plain words
What is it for?
Use it to install and configure messaging tools, then support requests to list chats, read or search messages, send messages, or react to them.
Why use it?
It removes the manual setup work and explains which security approvals must be given. Messages are only read or sent when the user asks.

Skill for Claude CodeCodex

Part of the texting plugin — 5 skills, 1 MCP server shipped together

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/bowlofarugula/texting/texting-setup
Any agent
npx skills add bowlofarugula/texting --skill texting-setup
Clone the repo
git clone --depth 1 https://github.com/bowlofarugula/texting

Made for: Claude Code, Codex.

Or install texting, the plugin that ships this one along with the rest of its 5 skills, 1 MCP server.

Per session 69 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,159 The whole file, excluding the scripts and references it only reads on demand.
Security scan F 4 findings. 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 $0.00069 $0.02159
Opus 5 $0.00034 $0.01079
Sonnet 5 $0.00014 $0.00432
Haiku 4.5 $0.00007 $0.00216

Measured 2d ago against content hash 830e39237da7, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade F, and why

texting-setup scanned grade F with 4 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 2d 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.

Downloads and executes remote codehighSupply chain

curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.

Missing → `curl -fsSL https://bun.sh/install | bash`. The plugin's `start`

Reads agent configuration directoriesmediumAgent snooping

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

`~/.claude/settings.json` (read first; preserve everything else):

Recursive force deletehighDestructive command

rm -rf with a variable or a broad path is one typo away from removing the wrong tree.

rm -rf "$HOME/.claude/texting/engine"

Makes network callslowCapability

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

tools; macOS's built-in `curl`/`unzip` do the work. First check whether it's
skills/texting-setup/SKILL.md · 189 lines

How it starts

The opening of the file, as written. The whole thing — 189 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/texting-setup — iMessage connector setup

After this runs, the user can say "text Sam we're confirmed for Friday" or "any new messages from Sam?" in any Claude session, with normal permission prompts before anything is sent. There is no background agent and no auto-reply: Claude only reads or sends when asked.

The macOS iMessage/SMS engine is imsg; the plugin ships a thin MCP server over it that exposes send_message, list_chats, read_messages, search_messages, and react in every session.

The user may be non-technical. Run every check yourself via Bash; only hand them the genuinely-human steps (System Settings toggle, macOS prompts), one at a time, in plain language.

Arguments passed: $ARGUMENTS (optional)


Step 1 — install the imsg engine (you do this, via Bash)

The engine is the official signed and notarized macOS binary from the openclaw/imsg GitHub release. You download and install it — the user does nothing and needs no Homebrew or developer tools; macOS's built-in curl/unzip do the work. First check whether it's already there:

"$HOME/.claude/texting/engine/imsg" --version 2>/dev/null || imsg --version 2>/dev/null || echo MISSING

A version number → done, skip to Step 2. (imsg is macOS-only.)

MISSING → install it (pinned version, checksum-verified):

set -e
IMSG_VERSION="v0.11.1"
IMSG_SHA256="a25a541f0c4c8244f301a8495f875964dc4b0c3fd5cbf5ead6a64e4d282d940e"
TMP="$(mktemp -d)"
curl -fsSL -o "$TMP/imsg-macos.zip" \
  "https://github.com/openclaw/imsg/releases/download/$IMSG_VERSION/imsg-macos.zip"
echo "$IMSG_SHA256  $TMP/imsg-macos.zip" | shasum -a 256 -c -
mkdir -p "$HOME/.claude/texting"
rm -rf "$HOME/.claude/texting/engine"
unzip -q "$TMP/imsg-macos.zip" -d "$HOME/.claude/texting/engine"
rm -rf "$TMP"
"$HOME/.claude/texting/engine/imsg" --version

A version number at the end → installed. If the checksum line fails, stop — do not install the file — and re-download once; persistent mismatch means the download is corrupted or tampered with, report it instead of proceeding.

Read the full file on GitHub · 189 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. 2d ago First seen · 189 lines · 69 tokens per session scan F 830e39237da7

Subscribe to this mod's changes

texting-setup is a skill published in the GitHub repository bowlofarugula/texting (1 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 69 tokens to every session and 2,159 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it F with 4 findings (downloads and executes remote code, reads agent configuration directories, recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.