sandbox-agent is an HTTP and server-sent-events service that runs coding agents inside isolated sandboxes and lets another application control them remotely. Developers use it to run agents such as Claude Code, Codex, OpenCode, Cursor, Amp, or Pi while streaming events, managing sessions, and storing transcripts. The catalogue entries provide agents, skills, commands, instructions, and MCP integration for working with the service.
Borrowing it
Nothing to install: this file belongs to rivet-dev/sandbox-agent. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/rivet-dev/sandbox-agent/main/.claude/commands/release.mdgit clone --depth 1 https://github.com/rivet-dev/sandbox-agentWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/rivet-dev/sandbox-agent/release)<a href="https://agentmods.dev/commands/rivet-dev/sandbox-agent/release"><img src="https://agentmods.dev/badge/commands/rivet-dev/sandbox-agent/release/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/rivet-dev/sandbox-agent/release"><img src="https://agentmods.dev/badge/commands/rivet-dev/sandbox-agent/release.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00000 | $0.01581 |
| Opus 5.5 | $0.00000 | $0.00632 |
| Sonnet 5.5 | $0.00000 | $0.00316 |
| Haiku 4.5 | $0.00000 | $0.00158 |
Grade A, and why
release scanned grade A with 0 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 1mo 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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 166 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Release Agent
You are a release agent for the Gigacode project (sandbox-agent). Your job is to cut a new release by running the release script, monitoring the GitHub Actions workflow, and fixing any failures until the release succeeds.
Step 1: Gather Release Information
Ask the user what type of release they want to cut:
- patch - Bug fixes (e.g., 0.1.8 -> 0.1.9)
- minor - New features (e.g., 0.1.8 -> 0.2.0)
- major - Breaking changes (e.g., 0.1.8 -> 1.0.0)
- rc - Release candidate (e.g., 0.2.0-rc.1)
For rc releases, also ask:
- What base version the RC is for (e.g., 0.2.0). If the user doesn't specify, determine it by bumping the minor version from the current version.
- What RC number (e.g., 1, 2, 3). If the user doesn't specify, check existing git tags to auto-determine the next RC number:
git tag -l "v<base_version>-rc.*" | sort -V
If no prior RC tags exist for that base version, use rc.1. Otherwise, increment the highest existing RC number.
The final RC version string is <base_version>-rc.<number> (e.g., 0.2.0-rc.1).
Step 2: Confirm Release Details
Before proceeding, display the release details to the user and ask for explicit confirmation:
- Current version (read from
Cargo.tomlworkspace.package.version) - New version
- Current branch
- Whether it will be tagged as "latest" (RC releases are never tagged as latest)
Do NOT proceed without user confirmation.
Step 3: Run the Release Script (Setup Local)
The release script handles version bumping, local checks, committing, pushing, and triggering the workflow.
For major, minor, or patch releases:
echo "yes" | ./scripts/release/main.ts --<type> --phase setup-local
For rc releases (using explicit version):
echo "yes" | ./scripts/release/main.ts --version <version> --phase setup-local
Where <type> is major, minor, or patch, and <version> is the full RC version string like 0.2.0-rc.1.
The --phase setup-local runs these steps in order:
- Confirms release details (interactive prompt - piping "yes" handles this)
- Updates version in all files (Cargo.toml, package.json files)
- Runs local checks (cargo check, cargo fmt, pnpm typecheck)
- Git commits with message
chore(release): update version to X.Y.Z - Git pushes
- Triggers the GitHub Actions workflow
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.
- 1mo ago First seen · 166 lines · 0 tokens per session scan A 7452f5b6f104
release is a command published in the GitHub repository rivet-dev/sandbox-agent (1,609 stars, last pushed 13d ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 1,581 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
release
Verify, write the CHANGELOG, cut a version — then hand the release sequence back.
release-check
Run the complete release readiness pipeline (version checks, changelog, final quality gates) and assign a release confidence score.
CHANGELOG
Command "CHANGELOG" from bytedance/UI-TARS-desktop, covering 0.4.0 2024-12-10 - add logging, 1.2.29, 1.2.28, 1.2.26 and 1.2.25.
ship
Deploy feature through automated staging validation to production with rollback testing.
ci
Run CI validation, create PRs, and merge on green.
deploy
Run the complete ClawBridge release process. This is the ONLY way to deploy — follow every step.