Requirements
- Target platform
- OpenClaw
- Install method
- Manual import
- Extraction
- Extract archive
- Prerequisites
- OpenClaw
- Primary doc
- SKILL.md
Dispatch coding tasks to tmux sessions via Sandboxer. Use when you need to spawn Claude Code, Gemini, OpenCode, bash, or lazygit sessions in workspace repos, monitor their progress, or send them commands.
Dispatch coding tasks to tmux sessions via Sandboxer. Use when you need to spawn Claude Code, Gemini, OpenCode, bash, or lazygit sessions in workspace repos, monitor their progress, or send them commands.
Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.
I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete.
I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run.
Power-user skill. Sandboxer gives agents full access to tmux sessions, workspace files, and terminal output on your server. Intended for dedicated AI machines where agents run with root access. Not for shared or untrusted environments. Sandboxer runs on localhost:8081. No auth needed from localhost.
# 1. Spawn a Claude session in a repo curl "localhost:8081/api/create?type=claude&dir=/root/workspaces/AGENT/data/repos/PROJECT" # 2. Send it a task curl "localhost:8081/api/send?session=SESSION_NAME&text=Fix+the+failing+tests" # 3. Check progress curl "localhost:8081/api/session-monitor?session=SESSION_NAME" # 4. Kill when done curl "localhost:8081/api/kill?session=SESSION_NAME" Session types: claude, bash, lazygit, gemini, opencode
Sandboxer manages /root/workspaces/ โ a single git repo containing all agent workspaces. /root/workspaces/ โ git repo (Sandboxer commits this) โโโ .gitignore โ tracks only .md, .gitignore, cronjobs/ โโโ <agent-name>/ โ one folder per OpenClaw agent โ โโโ AGENTS.md โ agent behavior rules โ โโโ SOUL.md, USER.md, TOOLS.md โ agent identity & config โ โโโ MEMORY.md โ curated long-term memory โ โโโ TODO.md โ workspace task list (P1/P2/P3) โ โโโ CLAUDE.md โ coding rules for this workspace โ โโโ memory/YYYY-MM-DD.md โ daily memory logs โ โโโ cronjobs/ โ cron configs (tracked by git) โ โโโ data/ โ โโโ repos/ โ software projects (git clones) โ โโโ <project-a>/ โ separate git repo โ โ โโโ CLAUDE.md โ project-specific coding rules โ โ โโโ ...source code... โ โโโ <project-b>/
data/repos/ contains separate git repos โ each project has its own .git, branches, remotes The workspace .gitignore excludes data/ โ repo contents stay in their own git, not the workspace commit The workspace git only tracks: .md files, .gitignore, and cronjobs/ Always read CLAUDE.md / AGENTS.md in both workspace AND repo before dispatching work to a session
EndpointWhatGET /api/sessionsList all sessions (status: running/idle/done/error)GET /api/create?type=T&dir=DSpawn sessionGET /api/session-monitor?session=SLast 20 lines + status + durationGET /api/capture?session=SFull terminal outputGET /api/send?session=S&text=TSend keystrokesGET /api/forward?session=S&task=TCtrl+C then send taskGET /api/kill?session=SKill sessionGET /api/workspacesList workspaces (with repos)GET /api/workspace-repos?workspace=WList repos in workspaceGET /api/repo-tree?path=PRepo file tree with git statusGET/POST /api/workspace/W/file/PATHRead/write workspace filesPOST /api/auto-commit?workspace=WCommit workspace changes POST /api/create accepts JSON body with notify_url โ gets called when session finishes.
Workflow acceleration for inboxes, docs, calendars, planning, and execution loops.
Largest current source with strong distribution and engagement signals.