Sssf Sandbox Orchestrator

What it does

When your agent uses it

• The user says mount a sandbox
• Run the factory in a sandbox
• Spin up N sandboxes
• Harvest a runs commits

Example prompts

• “/sssf-sandbox-orchestrator”

Requirements

• A credential in OPENROUTER_PROVISIONING_KEY

Workflow steps

6 steps, taken from the first numbered list in SKILL.md.

1. 1Never decide teardown. Report what the run produced, what it cost, and recommend — the human
2. 2Never run ADWs on the host. just adw sdlc "..." here runs the factory on the engineer's
3. 3A gate failure means STOP and diagnose, never destroy. The VM is left alive deliberately.
4. 4Never touch a key. The runtime key lives at .sandbox/runs/.key (0600) and inside the
5. 5Never mint outside create. The sbx- prefix is the entire safety model for reap; the
6. 6Report the run id every time. It is the only handle the next phase, the next session, and

What it can do on your machine

Read from SKILL.md and the folder at commit 92f1701. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  Shell commands in SKILL.md call:

  • just
  • ssh
  • git

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  No URLs in SKILL.md. Its commands use ssh and git, which can reach the network depending on how they are called.

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names these keys or tokens, usually read from environment variables:

  • OPENROUTER_PROVISIONING_KEY

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Sssf Sandbox Orchestrator loads about 2.6k tokens when it runs, and up to ~16k if it reads all its reference files. Until then it costs about 118 tokens; SKILL.md has 1,349 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~118

When it runs · the whole SKILL.md, loaded when a task matches

~2.6k

With references · SKILL.md plus every file in references/, read only if the agent opens them

~16k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

• NoteMentions a .env fileSKILL.md:89

  .61s, no auth), optional SHA pin, write `.env` with the runtime key |

• NoteMentions a .env fileSKILL.md:146

  VM's `app/.env`. Never print it, never copy it, never pass it over ssh — `setup`'s gate spends it

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from disler/inkwell-agent-sandboxes-and-software-factory at commit 92f1701, republished under its MIT licence (© disler). 1,349 words, ~2,617 tokens.

name

sssf-sandbox-orchestrator

description

Drive the six-phase sandbox mount system from the host — mount throwaway exe.dev VMs, run the Super Simple Software Factory inside them, watch from outside, harvest the commits, tear down. Use when the user says mount a sandbox, run the factory in a sandbox, spin up N sandboxes, best-of-N, check on a run, harvest a run's commits, or tear down. Keywords - sandbox, mount, exe.dev VM, run id, fan out, best-of-N, harvest, bundle, teardown, reap.

argument-hint

[mount|execute|agent|observe|harvest|teardown] [run-id or prompt]

just sbx manage doctor

just sbx
├── mount              the chain: create → fill → setup → observe
├── lifecycle          the six phases, for when you need one on its own
├── manage             preflight, readback, fleet ops — nothing here is a phase
├── run                put work in / look inside: `run cmd`, `run agent`
└── orch               boot a host-side orchestrator: `orch cc`, `orch pi`

© disler, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 13 other files (references) in .claude/skills/sssf-sandbox-orchestrator of disler/inkwell-agent-sandboxes-and-software-factory.

• SKILL.md
• cookbooks/access_a_running_sandbox.md
• cookbooks/debug_a_failed_gate.md
• cookbooks/execute_work.md
• cookbooks/fan_out_n.md
• cookbooks/just_command_model.md
• cookbooks/mount_one.md
• cookbooks/observe_and_report.md
• cookbooks/teardown_and_reap.md
• references/gotchas.md
• references/kickoff_paths.md
• references/models.md
• references/phases.md
• references/run_record.md

Open the folder on GitHubat commit 92f1701