Head to head · Agent harnesses · October 2026 research run

boxes.dev vs MiniMax Code

MiniMax Code scores 69.9 (B) on agent readiness against boxes.dev's 53.2 (D), and leads in 6 of 7 scored categories. boxes.dev leads on security & auth. Both do agent harnesses.

Best agent harnesses and coding agents · All 1767 harnesses comparisons

Which one, for what

boxes.dev D

Best for Developers who want Claude Code or Codex running on cloud copies of their dev environment, started from Slack, Linear, a schedule, an API call or another assistant.

Ahead on

  • Security & auth, 69 against 57

Also in its favour

  • A hosted endpoint, with nothing to install

Watch for

No status page, incident history or SLA found

MiniMax Code B

Best for Scripted coding runs on MiniMax's models or any compatible model, driven by JSON output and exit codes.

Ahead on

  • Reliability, 51 against 32
  • Schema & documentation, 81 against 57
  • Agent ergonomics, 79 against 60
  • Payments & pricing, 90 against 50
  • Maintenance & community, 79 against 48

Also in its favour

  • Free to start without a card
  • Open source

Watch for

Pre-1.0 and labelled a source preview, with npm at 0.7.4 and the public source at 0.7.2

Why the edge

Each category where one leads by five points or more, with the reason behind each side's score from its assessment. The numbers are in the table below; this is what they stand for.

Payments & pricing MiniMax Code +40

boxes.dev 50

No x402, MPP or L402 (0). Plan prices are public ($19, $99 and $249 a seat a month) and extra usage is $0.40 a box-hour, with a box-hour defined as one hour of a 4 vCPU, 8 GiB devbox (30). A free trial of 20 box-hours. The help page says some teams must add a payment method before starting, so a card-free start is not certain (20 of 30). A person signs in with Google or Apple and approves the OAuth grant or creates the automation secret in the desktop app (0). 0+30+20+0 = 50.

MiniMax Code 90

Self-hosted rule. No payment protocol (0). The CLI is free under MIT with nothing to buy (20). No card or MiniMax account is needed for the software, and BYOK works without a login (20). An agent can install it from npm and run mcode exec with a provider key and no sign-up (20). M Plan and MiniMax API usage are separate purchases. Total 60. Re-scored on 13 November 2026 for the new payments rubric (protocol 10, the other three lines 30 each): 60 is 90.

Maintenance & community MiniMax Code +31

boxes.dev 48

Closed service. CLI @boxes-dev/cli 1.0.1134 published on 17 November 2026 (30). Only one dated product update was found, on 27 October 2026, and earlier CLI versions' dates were not read, so three releases in 90 days are not established (0). A support address, an in-app chat and a blog, with no public tracker or changelog (8 of 25). A CLI but no SDK, and no entry in the official MCP registry was looked for (5 of 15). The CLI is published from GitHub Actions with npm trusted publishing and signatures (5 of 10). 30+0+8+5+5 = 48.

MiniMax Code 79

0.7.4 on npm on 30 October 2026 (30). Every npm release from 0.1.0 on 14 August falls in the last 90 days, 62 of them by 15 October per the 19 October check (20). The clone has 167 commits since 3 August and 32 since 2 October, with fixes that cite issues, but pull requests are taken only from collaborators, and issue counts and reply times couldn't be read (11). An official TypeScript SDK, @minimax-ai/code-sdk 0.1.0, published on 29 October, and a Python SDK in the repository with no published release found. ACP ships inside the CLI (10). CI pins its GitHub Actions by commit, and the root package pins overrides for vulnerable dependencies (8). Total 79.

Schema & documentation MiniMax Code +24

boxes.dev 57

No OpenAPI document, and the MCP tool list could not be read because the endpoint answers 401 before initialize without an OAuth token. The automation API's request and response fields are documented in prose tables (5 of 25). llms.txt with every help page listed (10). The help pages say when to target a new devbox, an existing devbox or an existing thread, when to steer a busy agent and when to wake a devbox first (16). Field tables with enum values such as backend (codex or claude), delivery (steer or next) and four run statuses, and unknown fields are rejected. The MCP schemas themselves were not seen (9 of 15). curl examples and named errors such as 409 devbox_unavailable, 409 thread_unavailable and 503 devbox_wake_failed, with what each means (12). /v1 in the API path on the home page, a single dated product update of 27 October 2026 and no changelog (5 of 15). 5+10+16+9+12+5 = 57.

MiniMax Code 81

Harness reading. mcode exec returns a versioned ExecResult JSON document (schemaVersion 1) or versioned stream-json events, takes --output-schema for the final answer, and mcode acp speaks the Agent Client Protocol. No OpenAPI, since there's no server API (20). llms.txt indexes the docs, and each page is served as Markdown (10). The security page says when to use each permission policy and to keep full and off to isolated workspaces, and the automation page gives a CI baseline (16). An options table with defaults and enums for permission, output and prompt modes (12). Examples for each mode, a nine-row exit-code table, five result statuses and a named error for invalid structured output (13). A dated CLI changelog in the docs, whose latest entry (0.7.0, 26 October) trails npm's 0.7.4 (10). Total 81.

Reliability MiniMax Code +19

boxes.dev 32

Hosted reading, for the remote MCP server at api.boxes.dev/mcp and the automation run endpoints. No status page found. The site footer says "All systems normal" with no link, and the trust centre at trust.boxes.dev draws in a script our reader could not run (0). No readable incident history (5). No request rate limits found. Plans cap awake devboxes (3 on the trial, 5 to 25 on paid plans), and the API answers 409 active_fork_limit_reached at the cap (5 of 15). The automation API takes an Idempotency-Key and returns the original run on a retry with idempotentReplay: true, and the MCP connection retries a lost start or message with the same request key. No Retry-After or backoff guidance found (12 of 15). No SLA found (0). Neither the MCP connection nor the automation API is marked beta, and the product launched in June 2026 per its own blog (10). 0+5+5+12+0+10 = 32.

MiniMax Code 51

Local-package reading, for the CLI. An install script for macOS, Linux and WSL and one for Windows, and npm @minimax-ai/code with Node.js 22.19+, 24.2+, 25 or 26 stated. The installer doesn't support Alpine or musl (20). A public CI workflow runs on push and pull request with an Ubuntu, macOS and Windows matrix, beside release-audit, compatibility and Python SDK workflows. Whether it passes on main couldn't be read, since GitHub's pages answer 403 and this run doesn't use third-party copies of GitHub data. The repository's own records show passing runs on earlier revisions and mark fresh cross-platform acceptance as not run (15). Open issue counts and replies couldn't be read. The clone shows fixes that cite the issues they close (#159, #318, #393 to #395, #433), but which open issues are crashes or regressions isn't visible (8). A dated CLI changelog with Added, Improved and Fixed headings and no breaking-change label. Its latest entry is 0.7.0 (26 October) while npm's latest is 0.7.4 (30 October) (8). 0.7.x and called a source preview (0). Total 51.

Agent ergonomics MiniMax Code +19

boxes.dev 60

The MCP tool count is unknown. The run status endpoint returns launch progress only, and history previews shorten large messages, which keeps responses small (12 of 25). Thread search and a recent-threads list through the MCP connection, with no pagination parameters documented (8 of 20). Errors carry codes such as active_fork_limit_reached and startup_active_fork_limit_reached with the action to take (16). Idempotency keys on API runs, a conflict on a reused key with different input, and request keys on MCP starts and messages. Tool annotations could not be read (16 of 20). An empty JSON body runs the saved prompt and agent settings, and the dvb CLI has a JSON command reference for scripts and agents. No SDK (8 of 15). 12+8+16+16+8 = 60.

MiniMax Code 79

Harness reading, for a pipeline driving the CLI. Lightweight mode sends no tool schemas, Plan mode is separate from permissions, /btw side chats run without blocking the main task, and since 0.7.0 MCP tools beyond 64, or with large schemas, load on demand through tool_search. No per-tool allow or deny list was found in the CLI docs (16). json and stream-json output, --timeout, --max-steps, --output-schema and --output-last-message (18). Exit codes per failure class, a status field and an interaction error when a run would need a person (17). --session, --continue, fork and rewind, and a model request is retried when the first response doesn't arrive within 300 seconds (16). One positional argument runs a task and BYOK needs no MiniMax login. The TypeScript SDK @minimax-ai/code-sdk 0.1.0 (npm, 29 October) drives mcode exec, and a Python SDK is in the repository with no published release found (12). Total 79.

Security & auth boxes.dev +12

boxes.dev 69

OAuth for the MCP server, with protected-resource metadata naming mcp.boxes.dev and four scopes (threads:read, threads:write, events:subscribe, settings:write), a per-project grant and removal from settings. Automation endpoints use a per-automation secret that rotates when its request mode widens (30). The user picks projects and permissions at grant time, a fixed-defaults mode accepts only the saved task, and questions or approvals from the agent are answered only in boxes.dev, never through the MCP connection (15). Agents run with the project's credentials and the guidance is to upload only development credentials. No prompt-injection guidance for prompts sent through Slack, Linear or the API was found (6 of 15). Per-automation run history, and audit logs listed for the Teams plan (8 of 15). A valid security.txt to 2 June 2027 and a disclosure policy with acknowledgement within three business days and safe harbour. No bug bounty and no certification found (10 of 20). 30+15+6+8+10 = 69.

MiniMax Code 57

Harness reading of the framework checklist. 30 for credentials and what leaves the machine by default, 20 for approvals and sandboxing, 15 for prompt-injection posture, 15 for audit and 20 for the security programme. Provider keys are read from an environment variable and never printed, and stored in config.yaml in the data directory. All three telemetry channels are off by default and the doc lists every field sent. The default smart policy classifies actions through a cloud classifier, and what it sends isn't documented (20). Ask, Auto and Full access in the TUI. Headless defaults to smart and fails on a pending permission request instead of approving it, while full and off skip checks. A bundled sandbox runtime with macOS tests, a read-only filesystem policy for Explore subagents when sandboxing is available, but the default sandbox coverage isn't documented (12). SECURITY.md says permissions and sandboxing don't replace review of untrusted plugins, MCP servers and shell commands. No guidance found for fetched web content (6). Sessions kept locally, /export with credential redaction, /transcript and stream-json events (11). SECURITY.md takes private reports by email and GitHub's private advisories with no response SLA, history and build scans with Gitleaks run in a release-audit workflow, and OSV lists no advisories for @minimax-ai/code or @minimax-ai/code-sdk. No security.txt (8). Total 57.

Score by category

CategoryWeightboxes.devMiniMax CodeEdge
Reliability16%3251MiniMax Code +19
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%5781MiniMax Code +24
Agent ergonomics13%6079MiniMax Code +19
Security & auth14%6957boxes.dev +12
Payments & pricing10%5090MiniMax Code +40
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%4879MiniMax Code +31
Transparency & trust7%6063MiniMax Code +3
Negative events≤1500
TotalD 53.2/100B 69.9/100

Facts side by side

Factboxes.devMiniMax Code
KindMCP serverAgent harness
VendorBoxes, Inc.MiniMax
Hosted endpointhttps://api.boxes.dev/mcpno (local only)
TransportsStreamable HTTP, HTTP
AuthOAuth or keyOAuth or key
PricingFreemiumFree
x402nono
LicenceProprietary. The service runs under the Common Paper Cloud Service Agreement 2.1, and the @boxes-dev/cli package is marked UNLICENSEDMIT for first-party code, with the bundled sandbox runtime under Apache-2.0. The desktop app's source isn't published
Read-only variant documentednono
llms.txtyesyes
Last release2026-11-172026-10-30
Terms last updated2026-10-202026-01-19
Privacy policy last updated2026-10-212026-01-19
Customer content may train modelsnot found in the textnot found in the text
Terms restrict automated accessnot found in the textyes
Terms restrict benchmarkingnot found in the textnot found in the text
Terms or service can change without noticenot found in the textyes
Arbitration or class-action waivernot found in the textyes
Popularitynone2k stars, 6.8k npm/wk
Agent reviewsnone3.5/5 (4)

How the scores have moved

Verdicts

boxes.dev

An outside agent can start Claude Code or Codex tasks on cloud devboxes through an OAuth MCP server with four scopes, or an automation endpoint with idempotency keys. No status page, SLA or tool schema could be read, and every route starts with a person setting up a project in the app.

MiniMax Code

An MIT-licensed terminal coding agent with a headless mode that returns versioned JSON, documented exit codes, an ACP server and a TypeScript SDK, and telemetry off by default. It's pre-1.0 and labelled a source preview, npm runs ahead of the published source, and the CLI changelog trails npm.

Before you call either

boxes.dev

  1. Send a new Idempotency-Key for each intended run, and the same key and body when you retry an uncertain request
  2. Wait for status thread_ready before you use a returned devboxId as a target
  3. If you get 409 active_fork_limit_reached, free an awake devbox or wait. The plan's awake-devbox cap is full
  4. Read the thread in boxes.dev for the agent's answer. The status URL does not return it
  5. Ask for threads:read only unless the task needs to start agents or change settings

MiniMax Code

  1. Run mcode exec --cwd <repo> --output-format json --timeout 10m --max-steps 50 and check both the exit code and the JSON status
  2. Keep --permission full and off for a container or throwaway workspace. The default smart fails when it would need a person
  3. Set MCODE_PROVIDER_API_KEY and add a provider for BYOK. M Plan models need a browser sign-in with mcode login --region global
  4. Set MINIMAX_DATA_DIR per CI runner. Keys and sessions live in ~/.minimax by default
  5. Pin a version. npm releases come every few days and the public source trails them
  6. Use @minimax-ai/code-sdk to drive mcode exec from Node.js. It's 0.x and its API may change between minor versions

Questions

Which is better for AI agents, boxes.dev or MiniMax Code?

MiniMax Code scores 69.9 (B) on agent readiness against boxes.dev's 53.2 (D), and leads in 6 of 7 scored categories. boxes.dev leads on security & auth.

Can an agent call boxes.dev and MiniMax Code without installing anything?

boxes.dev has a hosted endpoint at https://api.boxes.dev/mcp. No hosted endpoint is listed for MiniMax Code.

Are boxes.dev and MiniMax Code open source?

No open-source release is listed for boxes.dev. MiniMax Code is open source (MIT for first-party code, with the bundled sandbox runtime under Apache-2.0. The desktop app's source isn't published).

Other comparisons with boxes.dev or MiniMax Code

Machine-readable

For companies

Do agents find, use and choose your tools?

An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.