Claude Code Task

Run coding tasks using a persistent tmux session with git worktree isolation. Supports multiple coding agents (Claude Code, Codex, CodeBuddy, OpenCode, etc.)...

Security Scan

The skill mostly matches its stated purpose (tmux + git worktree code tasks) but includes instructions that can expose local secrets, bypass agent permissions, and write files into your repo without declaring those privileges, so it needs careful review before use.

Purpose & Capability

The name/description match the instructions for using tmux and git worktrees. However, several requested behaviors (always symlinking .env files into worktrees, writing a MEMORY.md preference file) are not strictly necessary for a coding-task relay and broaden the skill's access to local secrets and repository files.

Instruction Scope

SKILL.md instructs the agent to create worktrees, start tmux sessions, capture tmux panes, send arbitrary keystrokes (including suggested automated 'y' and Ctrl+C in troubleshooting), and to symlink .env and .env.local into worktrees. These actions read and surface local configuration and can cause automatic interaction with CLIs; they go beyond merely relaying messages and could expose secrets or approve prompts without explicit user interaction.

Install Mechanism

Instruction-only skill with no install spec or downloaded code, which minimizes direct supply-chain risk. It does assume external binaries (tmux, git, node, and the chosen coding agent CLI) are present, but does not install anything itself.

Credentials

The skill declares no required env or credentials, yet explicitly symlinks project .env and .env.local into each worktree—this accesses local secret files without declaring or justifying credential access. It also instructs writing preferences to MEMORY.md (modifying repository or workspace files). These are disproportionate to what the metadata states.

Persistence & Privilege

The skill is not force-enabled (always:false) and can be invoked by the user. It does request persistent presence in the workspace by creating branches/worktrees and by writing a MEMORY.md preference. Modifying the repo (adding branches, saving memory/preferences) is within its purpose but should be made explicit to users because it alters repository state.

What to consider before installing

This skill is coherent with its stated goal of running coding agents in tmux + git worktrees, but it asks the agent to symlink your .env files into task worktrees and to write a MEMORY.md preference — both of which can expose secrets or modify your repo without obvious consent. It also suggests launching Claude with --dangerously-skip-permissions and contains troubleshooting snippets that may auto-send 'y' or Ctrl+C to the running agent. Before installing: (1) confirm you trust the coding agent CLIs you will run, (2) avoid or change the automatic .env symlink behavior if your .env contains secrets (prefer copying with explicit user consent or not exposing it), (3) verify where MEMORY.md will be written (agent memory vs repo) and whether you want that file created, (4) do not use the --dangerously-skip-permissions flag unless you understand its implications, and (5) consider running the workflow in a disposable or containerized environment first. If possible, ask the skill author to make .env symlinking optional and to remove any instructions that auto-approve prompts.

Coding Agent Task (tmux + worktree)

Run coding tasks by spawning a coding agent in a tmux session + git worktree. Every task gets its own isolated branch and persistent conversation.

Step 0: Determine which agent to use

Check memory first:

memory_search("preferred coding agent tool")

• If found → use that tool, no need to ask

• If not found → ask the user:

  "Which coding agent should I use? (default: claude) Options: claude, codex, opencode, codebuddy, or any CLI tool name"

  Then save the answer to memory:

  memory: preferred_coding_agent = <tool>
  

  Write to MEMORY.md under a "Preferences" section.

Default if user doesn't answer: claude

Step 1: Setup worktree

# Always use a worktree — one per task
git -C <project> worktree add -b <branch> <worktree-path> main

# Symlink env files
ln -sf <project>/.env <worktree-path>/.env
ln -sf <project>/.env.local <worktree-path>/.env.local   # if exists

Step 2: Start tmux session with the chosen agent

tmux new-session -d -s <task-name> -c <worktree-path>

Then launch based on tool:

Tool	Command
claude	claude --dangerously-skip-permissions
codex	codex
opencode	opencode
codebuddy	codebuddy (or check its CLI name)
other	use the tool's interactive CLI command

tmux send-keys -t <task-name> "nvm use 20 && <tool-command>" Enter

Step 3: Send task with plan-first instruction

tmux send-keys -t <task-name> -l -- "Your task here.

Before making any changes, show me a plan of what you intend to do and wait for my approval."
sleep 0.1
tmux send-keys -t <task-name> Enter

Step 4: Relay plan to user

# Poll for plan output
tmux capture-pane -t <task-name> -p | tail -30

When agent outputs a plan → relay it to the user, wait for their confirmation before proceeding.

Relay flow:

1. Agent outputs plan → relay to user
2. User says "ok" / requests changes → forward to agent
3. Agent proceeds → monitor and relay further questions

# Send user's response
tmux send-keys -t <task-name> -l -- "<user response>"
sleep 0.1
tmux send-keys -t <task-name> Enter

# Check if waiting for input
tmux capture-pane -t <task-name> -p | tail -10 | grep -E "❯|Yes.*No|proceed|permission|plan|approve"

Step 5: Parallel tasks

Same pattern, multiple sessions:

tmux new-session -d -s task-a -c /tmp/task-a
tmux new-session -d -s task-b -c /tmp/task-b

Check all at once:

for s in task-a task-b; do
  echo "=== $s ==="
  tmux capture-pane -t $s -p 2>/dev/null | tail -5
done

Step 6: Cleanup

git -C <project> worktree remove <worktree-path>   # branch preserved
tmux kill-session -t <task-name>

# User can then test in main workspace:
# git switch <branch>

Rules

• Check memory first — never ask for tool preference if already saved
• Always use worktrees — one per task, no exceptions
• Always use tmux — persistent session, multi-turn conversation
• Always show plan first, wait for user approval before agent touches files
• Always symlink .env files — don't copy
• One status message when starting, one when done or stuck
• See references/troubleshooting.md for common issues