Head to head · Agent harnesses · October 2026 research run

boxes.dev vs Muse Code

Muse Code scores 57.4 (C) on agent readiness against boxes.dev's 53.2 (D), and leads in 4 of 7 scored categories. boxes.dev leads on payments & pricing. 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

  • Payments & pricing, 50 against 30

Also in its favour

  • A hosted endpoint, with nothing to install
  • No incidents deducted, where Muse Code loses 5 points for them

Watch for

No status page, incident history or SLA found

Muse Code C

Best for Developers who want Meta's Muse models, Muse Wonder Full included, in a terminal or CI agent with the sandbox and approvals on from the first run.

Ahead on

  • Reliability, 45 against 32
  • Schema & documentation, 84 against 57
  • Agent ergonomics, 79 against 60
  • Maintenance & community, 75 against 48

Watch for

Closed source, with no public CI or issue tracker for the CLI found and release notes that carry no dates

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.

Schema & documentation Muse Code +27

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.

Muse Code 84

Framework reading. muse schema prints the JSON schema of a versioned session protocol, the protocol declarations and a stable manifest are published in the SDK repository, and muse exec --json emits JSONL events. No JSON Schema for settings.json was found (22 of 25). llms.txt at dev.meta.ai/llms.txt with a Markdown copy of each page (10). The permissions page says what each profile and flag is for and when to use it, such as Unrestricted only in an isolated environment, and the extending page says when to use subagents and when not to (16 of 20). Profiles, approval modes, sandbox network modes and reasoning-effort levels are enumerated, with MCP servers typed by transport (12 of 15). Examples on every page, exit codes for muse exec, and named settings errors, though no full list of run errors (12 of 15). A changelog for every version to 1.4.13, with no dates (12 of 15). 22 + 10 + 16 + 12 + 12 + 12 = 84.

Maintenance & community Muse Code +27

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.

Muse Code 75

1.4.13, the newest entry in the changelog, was mirrored into the SDK repository on 24 October 2026, and the SDKs moved to 1.4.14 there on 27 October (30). The SDK repository's dated release mirrors since 4 August are 1.3.0 (16 and 17 September), 1.3.1 (21 September), 1.4.2 (30 September), 1.4.4 (8 October), 1.4.6 (11 October), 1.4.7 (13 October), 1.4.8 (14 October), 1.4.9 (15 October), 1.4.10 (16 October), 1.4.11 (19 October), 1.4.12 (22 October), 1.4.13 (24 October) and 1.4.14 (27 October) (20). No public tracker for the CLI. The SDK repository takes issues with a written triage guide, and maintainers merged pull requests up to #132 on 30 October; issue replies unread (8 of 25, per the 12 October check). MIT TypeScript and Python SDKs in lockstep, whose READMEs now say the stable surface evolves additively with a documented deprecation path, while the docs keep the developer preview banner (12 of 15). A checksummed launcher, but no public CI for the program (5 of 10). 30 + 20 + 8 + 12 + 5 = 75.

Payments & pricing boxes.dev +20

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.

Muse Code 30

Harness reading of the published rubric. No payment protocol (0). Plan prices ($5, $15, $50 and $200 a month) and Model API token prices are public without a login, with plan usage given as multiples of the Everyday plan's 10 to 50 prompts every 5 hours (20). No free plan or trial found, and pay-as-you-go needs a card on the Model API account (0). A person signs in through a browser or creates a key in the dashboard, and no local-model mode is documented (0). 0 + 20 + 0 + 0 = 20. Re-scored on 13 November 2026 for the new payments rubric (protocol 10, the other three lines 30 each): 20 is 30.

Agent ergonomics Muse 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.

Muse Code 79

Framework reading, adapted to a harness driven by a pipeline. muse exec runs one prompt with the MCP servers in settings, and named toolsets select MCP tools by name or prefix. We found no deferred tool loading in Muse Code's own docs (18 of 25). --max-model-steps stops a stuck run, and muse export writes a session record, but no budget cap or output schema was found (15 of 20). Exit codes 0, 1, 2, 130 and 143 are documented, and retry notices carry the provider error and HTTP status (16 of 20). --session-id continues a headless run, muse export and the event log support replay, and rewind branches without changing the original (18 of 20). TypeScript and Python SDKs. The defaults (Auto-review, proxy-only network) can stop for a review nobody answers in CI, so a pipeline has to choose a posture (12 of 15). 18 + 15 + 16 + 18 + 12 = 79.

Reliability Muse Code +13

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.

Muse Code 45

Local-package reading. Official installers for macOS, Linux and Windows that fetch a native binary, and the install.sh script checks the launcher's SHA-256 against the advertised sum (20). No public CI or test suite for the program, since the source isn't published. The SDK repository's workflows only publish to npm, PyPI and crates.io, by hand (0). No public issue tracker for the CLI was found, so there's nothing to count either way. The SDK repository takes issues for the SDK and protocol only (5, a departure from the line, which assumes a tracker). Release notes per version under New, Improvements, Performance and Fixes, with security fixes labelled and one break (1.4.8's profile refusal) announced in the 1.4.6 notes, but no dates and no breaking-change heading (5). 1.4.13 (15). Same reading as Claude Code, which is also closed source. The 12 October reading of per-platform checksums in the release manifest is struck, since api.meta.ai's robots.txt now disallows every path. 20 + 0 + 5 + 5 + 15 = 45.

Score by category

CategoryWeightboxes.devMuse CodeEdge
Reliability16%3245Muse Code +13
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%5784Muse Code +27
Agent ergonomics13%6079Muse Code +19
Security & auth14%6967boxes.dev +2
Payments & pricing10%5030boxes.dev +20
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%4875Muse Code +27
Transparency & trust7%6056boxes.dev +4
Negative events≤150-5
TotalD 53.2/100C 57.4/100

Facts side by side

Factboxes.devMuse Code
KindMCP serverAgent harness
VendorBoxes, Inc.Meta
Hosted endpointhttps://api.boxes.dev/mcpno (local only)
TransportsStreamable HTTP, HTTP
AuthOAuth or keyOAuth or key
PricingFreemiumPaid
x402nono
LicenceProprietary. The service runs under the Common Paper Cloud Service Agreement 2.1, and the @boxes-dev/cli package is marked UNLICENSEDProprietary binary, used under the Meta Model API Terms of Service, which cover Meta's coding harness. Subscriptions are also under the Meta Subscription Terms of Service. The TypeScript and Python SDKs in the linked repository are MIT.
Read-only variant documentednoyes
llms.txtyesyes
Last release2026-11-172026-10-24
Terms last updated2026-10-202026-10-02
Privacy policy last updated2026-10-21couldn't be read
Customer content may train modelsnot found in the textyes
Terms restrict automated accessnot found in the textnot found in the text
Terms restrict benchmarkingnot found in the textyes
Terms or service can change without noticenot found in the textnot found in the text
Arbitration or class-action waivernot found in the textyes
Popularitynone63k npm/wk
Agent reviewsnone3.3/5 (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.

Muse Code

Muse Code runs with approvals and an OS sandbox on by default, per-host network approval, an append-only event log and a headless mode with documented exit codes. It ships often, but its release notes carry no dates, its security fixes get no advisories, and telemetry and Contributor mode, which lets Meta train on code, are on unless changed.

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

Muse Code

  1. Set META_API_KEY for headless runs. Browser sign-in isn't available in CI or for Meta Managed Accounts
  2. Gate on your own test command, not the exit code. muse exec returns 0 when the turn completes even if tests still fail
  3. Pass --max-model-steps on unattended runs so a stuck task can't loop
  4. Prefer --disable-approval to --yolo in CI so the sandbox stays on. On Linux the runner needs a working bubblewrap and a non-musl build
  5. Turn Contributor mode off in /settings before a subscription session touches private code

Questions

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

Muse Code scores 57.4 (C) on agent readiness against boxes.dev's 53.2 (D), and leads in 4 of 7 scored categories. boxes.dev leads on payments & pricing.

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

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

Other comparisons with boxes.dev or Muse 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.