Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsBlogPro
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Authors
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges
  • Chrome Extension
  • Skill Manager

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

ProTermsPrivacyRefunds
Back to skills

T3

ASecurity

Start, inspect, pair, or stop T3 Code in the AGRO sandbox, locally or over a private Tailscale tailnet for phone access. Use this for the headless `t3 serve` harness on port 3773, including tmux launch, preflight diagnosis, pairing URL discovery, minting a pairing URL for a second device, logs, status, and shutdown. T3 Code wraps an already-authenticated Claude Code, Codex, or OpenCode backend. TRIGGER when: user asks to run T3 Code, start `t3 serve`, reach T3 Code from a phone or another mac...

40 stars
0 votes
0 copies
0 views
Added 10/7/2026
ai-agentsgobashnodebackend

Works with

claude codeterminal

Security Analysis

A100/100

Pro scans all 4 files and shows the line behind each finding

Scanned 10/7/2026

$npx -y skills add mifunedev/agro --skill t3 --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of T3?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for T3
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/mifunedev-agro/badge)](https://www.skillsdirectory.com/skills/mifunedev-agro)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
Files
SKILL.md
---
name: t3
description: |
  Start, inspect, pair, or stop T3 Code in the AGRO sandbox, locally or
  over a private Tailscale tailnet for phone access. Use this for the headless
  `t3 serve` harness on port 3773, including tmux launch, preflight diagnosis,
  pairing URL discovery, minting a pairing URL for a second device, logs,
  status, and shutdown. T3 Code wraps an already-authenticated Claude Code,
  Codex, or OpenCode backend.
  TRIGGER when: user asks to run T3 Code, start `t3 serve`, reach T3 Code from a
  phone or another machine, pair a device, get the T3 pairing URL, diagnose why
  T3 Code or the tailnet will not start, check T3 Code status/logs, or stop
  T3 Code.
argument-hint: "[start|status|url|pair|logs|stop|attach|doctor|help] [--session <name>] [--port <port>] [--log <path>] [--tailscale] [--tailscale-port <p>]"
allowed-tools: Bash, Read
disable-model-invocation: true
---

# T3 Code

`npx t3` (no subcommand) is the desktop GUI launcher and is not what this skill
runs. Headless and remote access use `npx t3 serve`; a new device is added to an
already-running server with `npx t3 pair`.

## Arguments

Arguments received: `$ARGUMENTS`

- `ACTION`: optional first positional argument; default `start`
  - `start`: run the preflight, then start `t3 serve` in tmux, or report the existing session
  - `status`: show whether the tmux session is running and print recent output
  - `url`: print the latest pairing URL from the log/pane if present
  - `pair`: mint a fresh one-time pairing URL against the running server, without restarting it
  - `logs`: print recent log lines
  - `stop`: kill the tmux session
  - `attach`: print the attach command; do not attach from an agent run
  - `doctor`: run the preflight checks and print one actionable line per failure
  - `help`: print script usage
- `--session`: tmux session name; default `agent-t3code`
- `--port`: expected T3 Code port; default `3773`
- `--log`: log file; default `/tmp/<session>.log`
- `--tailscale`: publish over Tailscale Serve on the tailnet (`t3 serve --tailscale-serve`, `t3 pair --tailscale`)
- `--tailscale-port`: alternate Tailscale Serve HTTPS port; default `443`

## Preconditions

T3 pairing is **not** provider auth. Pairing a phone does not log any provider
in, and a provider login does not pair a device. They are two separate
credentials on two separate lifecycles.

At least one backend must already be installed and authenticated inside the
sandbox before T3 Code is useful:

```bash
claude        # complete OAuth on first launch
codex login
opencode auth login
```

T3 Code itself prints a single-use pairing URL such as
`http://localhost:3773/pair#token=...`, or an
`https://<machine>.<tailnet>.ts.net/...` URL in Tailscale mode. Treat that URL
and its token as a secret: never paste it into an issue, a PR, a tracked file,
or a persistent log.

The preflight (`doctor`, and the first step of `start`) checks:

- `tmux`, `npx`, and `node` on `PATH`
- Node satisfies `^22.16 || ^23.11 || >=24.10` (the T3 server's `engines.node`)
- with `--tailscale`: the `tailscale` binary is installed, `tailscaled` is
  running and reachable, and the tailnet backend state is `Running`
- the T3 port answers on loopback when a session is already up

## Run

Run the bundled script with the received arguments:

```bash
bash "$CLAUDE_SKILL_DIR/scripts/t3-code.sh" $ARGUMENTS
```

Launch commands the script emits, verbatim:

| Invocation | Command |
| --- | --- |
| `/t3 start` | `npx --yes t3 serve` |
| `/t3 start --tailscale` | `npx --yes t3 serve --tailscale-serve` |
| `/t3 start --tailscale --tailscale-port 8443` | `npx --yes t3 serve --tailscale-serve --tailscale-serve-port 8443` |
| `/t3 pair` | `npx --yes t3 pair` |
| `/t3 pair --tailscale` | `npx --yes t3 pair --tailscale` |

The server always runs under the sandbox tmux convention
(`tmux new-session -d -s agent-t3code '... 2>&1 | tee /tmp/agent-t3code.log'`),
so it survives a terminal disconnect. It stays bound to container loopback; the
skill never binds T3 Code to a public interface.

For the phone-side recipe and the tailnet session layout, read
[`references/tailscale-mobile.md`](references/tailscale-mobile.md). For the tmux
rules, read [`references/sandbox-processes.md`](references/sandbox-processes.md).

## Report

After `start`, report:

- tmux session name and log path
- the exact launch command that was used
- pairing URL if found, otherwise the command to inspect logs
- local URL, normally `http://localhost:3773`
- in Tailscale mode, the tailnet HTTPS port and that the URL is the node's
  MagicDNS name
- reminder: over SSH/remote host without a tailnet, use VS Code port forwarding
  or see `docs/connecting.md`
- revocation paths:
  - `t3 auth` — issue, inspect, and revoke T3 sessions and credentials
  - `tailscale serve --https=443 off` — withdraw the Serve mapping (it persists
    until you do)
  - `tailscale logout`, or delete the node in the Tailscale admin console —
    remove the device from the tailnet

For public sharing beyond a private tailnet, use `/cloudflared 3773` only after
confirming the operator wants a public bearer URL. Tailscale is private; a
Cloudflared tunnel is not.

Attribution

mifunedevmifunedev
View sourceSee grades on GitHubMore from mifunedev →
SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Related Skills

Caveman

Terse caveman voice: answer first, fluff gone, every technical fact kept. Use for /caveman, "caveman mode", "talk like caveman", "be brief", "less tokens". Stays on until "stop caveman" or "normal mode".

1100021 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

698431 votes

Writing Skills

Create and manage Claude Code skills in HASH repository following Anthropic best practices. Use when creating new skills, modifying skill-rules.json, understanding trigger patterns, working with hooks, debugging skill activation, or implementing progressive disclosure. Covers skill structure, YAML frontmatter, trigger types (keywords, intent patterns), UserPromptSubmit hook, and the 500-line rule. Includes validation and debugging with SKILL_DEBUG. Examples include rust-error-stack, cargo-dep...

3931 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3421 votes

catchup

Recovers the conversation and failed tool calls of a previous Codex, Amp, Claude Code, Antigravity, Cline, Copilot CLI, Cursor, DeepSeek Harness, Grok Build, Kimi, OpenCode, Pi Agent, or ZCode session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", asks to recover/summarize a previous session before continuing, or asks to diagnose or report a catchup failure. Do NOT use for the current conversation, git history, or any non-agent log.

741 votes
View all in ai-agents →