{
  "fixes": {
    "slug": "vercel-sandbox",
    "name": "Vercel Sandbox",
    "listing": "https://www.anchorterminal.com/tools/vercel-sandbox",
    "markdown": "# Fix list: Vercel Sandbox\n\nFrom Anchor Terminal's listing at https://www.anchorterminal.com/tools/vercel-sandbox, the October 2026 research run, assessed 1 October 2026. Grade B, 69.6 out of 100.\n\nThis is everything the published grade says the listing lacks, the biggest possible gain to the total first. It comes from the reason given for each score, the checklist each category was scored against (https://www.anchorterminal.com/benchmark/#checklist), the provenance checks, the deductions, what we couldn't check and what the review panel asked for. A fix counts at the next check, once it's public.\n\nFor a coding agent working on Vercel Sandbox: work through the items below in the product, its docs and its public pages. Each category gives the reason for its score, with the points each checklist item earned, and the checklist itself, so the gap is the items that earned less than their points. Change the product, not the wording, and keep a note of what you changed and where it's published.\n\n## 1. Payments \u0026 pricing, 40 out of 100, up to 7.5 more on the total\n\nWhy it scored 40: No x402, MPP or L402 (0). Per-unit prices published by region, $0.128 an Active CPU hour and $0.0212 a GB-hour of memory at iad1 (20). Hobby includes 5 Active CPU hours and 5,000 creations a month at no cost, but we didn't confirm this run that Hobby signup needs no card (10). Stripe Projects lists Vercel, so an agent can create the account through the operator's Stripe login (10).\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-payments):\n\nThe published rubric, also on the [x402 page](https://www.anchorterminal.com/x402/).\n\n- 40, a machine payment protocol (x402, MPP or L402) on the tool's own endpoints. 10 to 30 when it covers only some endpoints or only goes through a third party, and the note says which.\n- 20, per-call or per-unit pricing published without a login. 10 for public plan-only pricing, 0 for \"contact sales\" or prices behind a login.\n- 20, a free tier or trial that doesn't need a card.\n- 20, autonomous onboarding, meaning an agent can get access without a person signing up in a browser (keyless use, x402, a programmatic key API).\n\nPayment platforms and agent wallets rarely charge for their own API over a machine protocol, so the first line has steps for them, and the highest one that applies counts. 40 when x402, MPP or L402 runs on all their own endpoints, 30 when it runs on part of their own API, 25 when their merchants can accept one, 20 for running a facilitator, 15 for paying as a buyer, and 0 when the only protocol is their own. Merchant acceptance sits above a facilitator because the platform's own customers can charge agents through it, while a facilitator settles for sellers who wire up the protocol themselves. The counter-argument (a facilitator does more for the protocol as a whole) has a point. Each note says which step applied.\n\nOpen-source software you run yourself is scored on its hosted or paid option if it has one. A free, self-hosted package with nothing to buy gets 20, 20 and 20 for the last three lines, and 0 to 40 for the first only if it ships a payment protocol.\n\n## 2. Reliability, 70 out of 100, up to 6 more on the total\n\nWhy it scored 70: Sandbox is its own component on www.vercel-status.com, with an incident feed (20). The feed shows elevated latency on the Sandbox API for 1 hour 16 minutes on 4 September 2026 and a 45-minute dashboard observability incident that included Sandboxes on 23 July, both degradations rather than outages (20). Control-plane limits published per plan, 1,000 requests a minute on Hobby, 10,000 on Pro and 100,000 on Enterprise, with deletes at 20 a second (15). `Sandbox.getOrCreate` with a name gives a safe retry, but we found no 429 or Retry-After guidance (5). No SLA for Sandbox found (0). GA (10).\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-reliability):\n\nHosted APIs, MCP servers, models and platforms.\n\n- 20, a public status page with component history (Statuspage, Instatus, BetterStack or the vendor's own).\n- 0 to 30, the incident record for the last 90 days on that page. 30 for a clean record or trivial incidents only, 20 for minor incidents only, 10 for one major outage (an hour or more of a core API down, or errors across the board), 0 for several. 5 when there's no history we could read, and the note says so.\n- 15, rate limits documented with numbers.\n- 15, documented 429 or overload handling (Retry-After, backoff guidance), and idempotency keys or safe-retry guidance where writes are involved.\n- 10, an SLA published for any paid tier.\n- 10, the surface agents use is generally available, not beta or preview.\n\nLocal packages, SDKs, frameworks and stdio MCP servers.\n\n- 20, installs from an official package with supported runtimes stated.\n- 25, a public CI and test suite, passing on the default branch.\n- 0 to 25, open crash or regression issues relative to activity (25 for few and handled, 0 for many, old and unanswered).\n- 15, semver discipline and breaking changes called out in a changelog.\n- 15, version 1.0 or later, or declared stable.\n\nProtocols are read from their reference implementations, the public facilitators or servers, spec stability and test vectors.\n\n## 3. Agent ergonomics, 65 out of 100, up to 5.7 more on the total\n\nWhy it scored 65: Sandbox objects are small, with no field selection (15). Key-value tags for grouping, with custom tags in beta (10). We didn't find a documented error format (10). `Sandbox.getOrCreate` with a name lands retries in the same sandbox (15). JavaScript and Python SDKs and a CLI, a sandbox starts with no required parameters, and persistence is on by default (15).\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-ergonomics):\n\n- 0 to 25, context cost. For MCP, the number and size of the tool definitions (25 for ten or fewer compact tools, 15 for 11 to 30, 5 for more than 30, plus up to 10 back for toolsets, dynamic loading or read-only subsets). For APIs, whether responses can be sized (field selection, limits, summaries).\n- 20, pagination, filtering and output-size controls.\n- 20, actionable, documented error responses, codes and messages an agent can recover from.\n- 20, idempotency or safe retries, and for MCP the `readOnlyHint` and `destructiveHint` annotations.\n- 15, sensible defaults, few required parameters, and official SDKs in at least two languages.\n\nModels are read for tool use, structured output, prompt caching, context length, batch and SDKs. Frameworks for how much code and how many defaults a tool-calling agent with MCP needs.\n\n## 4. Schema \u0026 documentation, 77 out of 100, up to 3.7 more on the total\n\nWhy it scored 77: We didn't find sandbox endpoints in a public OpenAPI file, so the typed JavaScript and Python SDK references count as 15 of 25 (15). Every docs page is served as Markdown with front matter and a link map for agents (10). The docs explain when to use persistent or ephemeral sandboxes and how Active CPU billing works, with a worked cost example (15). Typed SDK parameters (12). KB guides and examples throughout, with no error reference found (10). Vercel's dated changelog carries Sandbox entries, and the SDKs follow semver (15).\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-schema):\n\nAPIs and MCP servers.\n\n- 25, a machine-readable contract (a public OpenAPI file or similar; for MCP, typed JSON Schema inputs on every tool).\n- 10, llms.txt or Markdown docs served for agents.\n- 0 to 20, descriptions that say what a tool is for, when to use it and when not to, read from the tool definitions in the source or the API reference.\n- 0 to 15, typed inputs with enums, constraints and required fields, and no free-form JSON blobs.\n- 0 to 15, examples and documented error responses.\n- 15, versioning and a public changelog.\n\nModels are read from the API reference, the OpenAPI file, llms.txt, the structured-output and tool-use docs and the model cards. Frameworks from docs a model can follow, typed interfaces, examples and the API reference.\n\n## 5. Security \u0026 auth, 80 out of 100, up to 3.5 more on the total\n\nWhy it scored 80: Vercel OIDC tokens tied to one project, which `vercel env pull` hands out for 12 hours, or access tokens elsewhere. Access tokens reach a whole team rather than one sandbox (25). Each sandbox is a Firecracker microVM with its own kernel (10). The firewall has allow-all (the default), deny-all including DNS, and custom rules by domain (matched on SNI) or CIDR, changeable at runtime without restarting processes (10). Credential brokering runs a proxy outside the sandbox that adds credentials to outbound headers and overwrites any the sandbox code tries to set (15). Sandboxes can be listed and connected to from the dashboard, and we didn't check audit logs this run (5). Valid security.txt pointing to HackerOne and responsible.disclosure@vercel.com, and Vercel claims SOC 2 Type II for Sandbox. No Sandbox advisories found (15).\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-security):\n\n- 0 to 30, the credential model. 30 for OAuth 2.1 with scopes, or scoped and revocable keys with rotation. 20 for plain revocable API keys. 10 for one all-powerful key. 10 off when a secret can travel in a URL query string as a documented option.\n- 0 to 20, read-only or least-privilege modes, and confirmation or approval for destructive actions.\n- 0 to 15, prompt-injection posture where the tool returns untrusted content (documented mitigations or guidance). A tool that returns no untrusted content gets 10.\n- 0 to 15, audit logs or per-call visibility for the operator.\n- 0 to 20, a security programme. security.txt or a disclosure policy, a bug bounty, SOC 2 or ISO 27001, advisories handled in public.\n\nModels are read for retention, whether API data trains models (and whether that's off by default), zero-retention options and certifications. Frameworks for telemetry defaults, approval hooks, guardrails and sandboxing.\n\n## 6. Transparency \u0026 trust, 75 out of 100, up to 2.2 more on the total\n\nMade of editorial 50, provenance 100.\n\nWhy it scored 75: The SDK and CLI are Apache-2.0. The platform is closed under terms dated 1 June 2026 (20). The docs state that snapshots expire 30 days after last use by default and that ephemeral sandboxes are removed after 14 days of inactivity. We didn't read the privacy policy or DPA this run (15). No deprecation policy for Sandbox found (5). 19 regions named, each with its own prices. No subprocessor list checked (10).\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-transparency):\n\n- 0 to 30, source availability and licence clarity. 30 for open source under an OSI licence, 15 for closed with clear terms, 0 for unclear terms.\n- 0 to 30, data handling and retention statements that agree with each other (privacy policy, DPA, retention periods, subprocessors).\n- 0 to 20, a deprecation policy or notices with dates.\n- 0 to 20, telemetry disclosed with an opt-out (local software), or subprocessors and data locations disclosed (hosted).\n\nThe other half of Transparency and trust is the provenance score, computed from checked facts (below). The category score is the mean of the two.\n\n## 7. Maintenance \u0026 community, 80 out of 100, up to 1.8 more on the total\n\nWhy it scored 80: @vercel/sandbox 3.3.0 on 2026-09-11 (30). Four SDK releases from 7 to 16 July 2026 alone, 2.5.0 to 2.7.0, before the move to 3.x (20). We didn't check issue replies on vercel/sandbox this run, so this rests on the public changelog (10). Current official JavaScript and Python SDKs (15). CI not checked (5).\n\nThe checklist (https://www.anchorterminal.com/benchmark/#checklist-maintenance):\n\n- 0 to 30, time since the last release, or the last published model or API change for a closed service. 30 within 30 days, 20 within 90, 10 within 180, 0 older.\n- 20, at least three releases or dated changelog entries in the last 90 days.\n- 0 to 25, responsiveness. Issues and pull requests answered on GitHub (the open issues and how recent the replies are). For closed services, a public changelog and a support or community channel that answers, 0 to 15.\n- 15, presence in the official MCP registry under a verified namespace (MCP servers), or current official SDKs (APIs and models).\n- 10, package health, current dependencies and CI.\n\nModels are read for deprecation notice periods and model churn rather than release counts.\n\n## What we couldn't check\n\nWhat we couldn't read counted as absent. Publishing it on a page a plain HTTP fetch can read (not only in a browser) lets the next check count it.\n\n- Whether Hobby signup needs a card. We didn't confirm it this run.\n- Whether the Sandbox endpoints appear in Vercel's public OpenAPI file.\n- Whether Sandbox calls show up in Vercel's audit log, and on which plans.\n- What changed between @vercel/sandbox 2.7.0 (16 July 2026) and 3.x. The GitHub releases page we saw stopped at 2.7.0.\n- The control-plane limits and the snapshot and inactivity retention periods come from last week's listing research on the pricing and persistent-sandboxes pages, which we didn't reload this run because of fetch limits.\n\n## Weaknesses\n\n- Tied to a Vercel team and project even when called from elsewhere, and access tokens reach the whole team\n- Hobby caps sessions at 45 minutes and pauses creation once the monthly allowance is spent\n- Snapshots keep the filesystem, not memory or running processes\n- Egress is allow-all until you set a policy\n- No MCP server for Sandbox and no public OpenAPI for its endpoints found\n\n## What costs an agent a turn today\n\nThe notes we give agents before they call it. Each one is a workaround an agent shouldn't need.\n\n- Call `sandbox.stop()` when the task is done. Memory bills until the session ends\n- Use `Sandbox.getOrCreate` with a name so retries land in the same sandbox\n- Set `networkPolicy` to `deny-all` for untrusted code. The default is allow-all\n- Put API keys in credential brokering rules, not in the sandbox environment\n- Pass `persistent: false` for one-off runs so no snapshot is stored or billed\n\n## What the review panel asked for\n\n- Document 429 and Retry-After\n- Publish a Sandbox SLA\n- per-sandbox scoped tokens\n\n## When it's done\n\nSend what changed and where it's published as a dispute (https://www.anchorterminal.com/builders/#disputes, or `POST https://www.anchorterminal.com/api/v1/contact` with `\"kind\": \"dispute\"`). Disputes are answered in public, and the listing is checked again by the same checklist. Paying for an audit or a listing claim changes nothing here.\n",
    "grade": "B",
    "score": 69.6,
    "assessed": "2026-10-01",
    "run": "October 2026 research run",
    "categories": [
      {
        "key": "payments",
        "name": "Payments \u0026 pricing",
        "score": 40,
        "maxGain": 7.5,
        "reason": "No x402, MPP or L402 (0). Per-unit prices published by region, $0.128 an Active CPU hour and $0.0212 a GB-hour of memory at iad1 (20). Hobby includes 5 Active CPU hours and 5,000 creations a month at no cost, but we didn't confirm this run that Hobby signup needs no card (10). Stripe Projects lists Vercel, so an agent can create the account through the operator's Stripe login (10).",
        "checklist": [
          "The published rubric, also on the [x402 page](https://www.anchorterminal.com/x402/).",
          "- 40, a machine payment protocol (x402, MPP or L402) on the tool's own endpoints. 10 to 30 when it covers only some endpoints or only goes through a third party, and the note says which.\n- 20, per-call or per-unit pricing published without a login. 10 for public plan-only pricing, 0 for \"contact sales\" or prices behind a login.\n- 20, a free tier or trial that doesn't need a card.\n- 20, autonomous onboarding, meaning an agent can get access without a person signing up in a browser (keyless use, x402, a programmatic key API).",
          "Payment platforms and agent wallets rarely charge for their own API over a machine protocol, so the first line has steps for them, and the highest one that applies counts. 40 when x402, MPP or L402 runs on all their own endpoints, 30 when it runs on part of their own API, 25 when their merchants can accept one, 20 for running a facilitator, 15 for paying as a buyer, and 0 when the only protocol is their own. Merchant acceptance sits above a facilitator because the platform's own customers can charge agents through it, while a facilitator settles for sellers who wire up the protocol themselves. The counter-argument (a facilitator does more for the protocol as a whole) has a point. Each note says which step applied.",
          "Open-source software you run yourself is scored on its hosted or paid option if it has one. A free, self-hosted package with nothing to buy gets 20, 20 and 20 for the last three lines, and 0 to 40 for the first only if it ships a payment protocol."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-payments"
      },
      {
        "key": "reliability",
        "name": "Reliability",
        "score": 70,
        "maxGain": 6,
        "reason": "Sandbox is its own component on www.vercel-status.com, with an incident feed (20). The feed shows elevated latency on the Sandbox API for 1 hour 16 minutes on 4 September 2026 and a 45-minute dashboard observability incident that included Sandboxes on 23 July, both degradations rather than outages (20). Control-plane limits published per plan, 1,000 requests a minute on Hobby, 10,000 on Pro and 100,000 on Enterprise, with deletes at 20 a second (15). `Sandbox.getOrCreate` with a name gives a safe retry, but we found no 429 or Retry-After guidance (5). No SLA for Sandbox found (0). GA (10).",
        "checklist": [
          "Hosted APIs, MCP servers, models and platforms.",
          "- 20, a public status page with component history (Statuspage, Instatus, BetterStack or the vendor's own).\n- 0 to 30, the incident record for the last 90 days on that page. 30 for a clean record or trivial incidents only, 20 for minor incidents only, 10 for one major outage (an hour or more of a core API down, or errors across the board), 0 for several. 5 when there's no history we could read, and the note says so.\n- 15, rate limits documented with numbers.\n- 15, documented 429 or overload handling (Retry-After, backoff guidance), and idempotency keys or safe-retry guidance where writes are involved.\n- 10, an SLA published for any paid tier.\n- 10, the surface agents use is generally available, not beta or preview.",
          "Local packages, SDKs, frameworks and stdio MCP servers.",
          "- 20, installs from an official package with supported runtimes stated.\n- 25, a public CI and test suite, passing on the default branch.\n- 0 to 25, open crash or regression issues relative to activity (25 for few and handled, 0 for many, old and unanswered).\n- 15, semver discipline and breaking changes called out in a changelog.\n- 15, version 1.0 or later, or declared stable.",
          "Protocols are read from their reference implementations, the public facilitators or servers, spec stability and test vectors."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-reliability"
      },
      {
        "key": "ergonomics",
        "name": "Agent ergonomics",
        "score": 65,
        "maxGain": 5.7,
        "reason": "Sandbox objects are small, with no field selection (15). Key-value tags for grouping, with custom tags in beta (10). We didn't find a documented error format (10). `Sandbox.getOrCreate` with a name lands retries in the same sandbox (15). JavaScript and Python SDKs and a CLI, a sandbox starts with no required parameters, and persistence is on by default (15).",
        "checklist": [
          "- 0 to 25, context cost. For MCP, the number and size of the tool definitions (25 for ten or fewer compact tools, 15 for 11 to 30, 5 for more than 30, plus up to 10 back for toolsets, dynamic loading or read-only subsets). For APIs, whether responses can be sized (field selection, limits, summaries).\n- 20, pagination, filtering and output-size controls.\n- 20, actionable, documented error responses, codes and messages an agent can recover from.\n- 20, idempotency or safe retries, and for MCP the `readOnlyHint` and `destructiveHint` annotations.\n- 15, sensible defaults, few required parameters, and official SDKs in at least two languages.",
          "Models are read for tool use, structured output, prompt caching, context length, batch and SDKs. Frameworks for how much code and how many defaults a tool-calling agent with MCP needs."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-ergonomics"
      },
      {
        "key": "schema",
        "name": "Schema \u0026 documentation",
        "score": 77,
        "maxGain": 3.7,
        "reason": "We didn't find sandbox endpoints in a public OpenAPI file, so the typed JavaScript and Python SDK references count as 15 of 25 (15). Every docs page is served as Markdown with front matter and a link map for agents (10). The docs explain when to use persistent or ephemeral sandboxes and how Active CPU billing works, with a worked cost example (15). Typed SDK parameters (12). KB guides and examples throughout, with no error reference found (10). Vercel's dated changelog carries Sandbox entries, and the SDKs follow semver (15).",
        "checklist": [
          "APIs and MCP servers.",
          "- 25, a machine-readable contract (a public OpenAPI file or similar; for MCP, typed JSON Schema inputs on every tool).\n- 10, llms.txt or Markdown docs served for agents.\n- 0 to 20, descriptions that say what a tool is for, when to use it and when not to, read from the tool definitions in the source or the API reference.\n- 0 to 15, typed inputs with enums, constraints and required fields, and no free-form JSON blobs.\n- 0 to 15, examples and documented error responses.\n- 15, versioning and a public changelog.",
          "Models are read from the API reference, the OpenAPI file, llms.txt, the structured-output and tool-use docs and the model cards. Frameworks from docs a model can follow, typed interfaces, examples and the API reference."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-schema"
      },
      {
        "key": "security",
        "name": "Security \u0026 auth",
        "score": 80,
        "maxGain": 3.5,
        "reason": "Vercel OIDC tokens tied to one project, which `vercel env pull` hands out for 12 hours, or access tokens elsewhere. Access tokens reach a whole team rather than one sandbox (25). Each sandbox is a Firecracker microVM with its own kernel (10). The firewall has allow-all (the default), deny-all including DNS, and custom rules by domain (matched on SNI) or CIDR, changeable at runtime without restarting processes (10). Credential brokering runs a proxy outside the sandbox that adds credentials to outbound headers and overwrites any the sandbox code tries to set (15). Sandboxes can be listed and connected to from the dashboard, and we didn't check audit logs this run (5). Valid security.txt pointing to HackerOne and responsible.disclosure@vercel.com, and Vercel claims SOC 2 Type II for Sandbox. No Sandbox advisories found (15).",
        "checklist": [
          "- 0 to 30, the credential model. 30 for OAuth 2.1 with scopes, or scoped and revocable keys with rotation. 20 for plain revocable API keys. 10 for one all-powerful key. 10 off when a secret can travel in a URL query string as a documented option.\n- 0 to 20, read-only or least-privilege modes, and confirmation or approval for destructive actions.\n- 0 to 15, prompt-injection posture where the tool returns untrusted content (documented mitigations or guidance). A tool that returns no untrusted content gets 10.\n- 0 to 15, audit logs or per-call visibility for the operator.\n- 0 to 20, a security programme. security.txt or a disclosure policy, a bug bounty, SOC 2 or ISO 27001, advisories handled in public.",
          "Models are read for retention, whether API data trains models (and whether that's off by default), zero-retention options and certifications. Frameworks for telemetry defaults, approval hooks, guardrails and sandboxing."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-security"
      },
      {
        "key": "transparency",
        "name": "Transparency \u0026 trust",
        "score": 75,
        "maxGain": 2.2,
        "reason": "The SDK and CLI are Apache-2.0. The platform is closed under terms dated 1 June 2026 (20). The docs state that snapshots expire 30 days after last use by default and that ephemeral sandboxes are removed after 14 days of inactivity. We didn't read the privacy policy or DPA this run (15). No deprecation policy for Sandbox found (5). 19 regions named, each with its own prices. No subprocessor list checked (10).",
        "blend": "editorial 50, provenance 100",
        "checklist": [
          "- 0 to 30, source availability and licence clarity. 30 for open source under an OSI licence, 15 for closed with clear terms, 0 for unclear terms.\n- 0 to 30, data handling and retention statements that agree with each other (privacy policy, DPA, retention periods, subprocessors).\n- 0 to 20, a deprecation policy or notices with dates.\n- 0 to 20, telemetry disclosed with an opt-out (local software), or subprocessors and data locations disclosed (hosted).",
          "The other half of Transparency and trust is the provenance score, computed from checked facts (below). The category score is the mean of the two."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-transparency"
      },
      {
        "key": "maintenance",
        "name": "Maintenance \u0026 community",
        "score": 80,
        "maxGain": 1.8,
        "reason": "@vercel/sandbox 3.3.0 on 2026-09-11 (30). Four SDK releases from 7 to 16 July 2026 alone, 2.5.0 to 2.7.0, before the move to 3.x (20). We didn't check issue replies on vercel/sandbox this run, so this rests on the public changelog (10). Current official JavaScript and Python SDKs (15). CI not checked (5).",
        "checklist": [
          "- 0 to 30, time since the last release, or the last published model or API change for a closed service. 30 within 30 days, 20 within 90, 10 within 180, 0 older.\n- 20, at least three releases or dated changelog entries in the last 90 days.\n- 0 to 25, responsiveness. Issues and pull requests answered on GitHub (the open issues and how recent the replies are). For closed services, a public changelog and a support or community channel that answers, 0 to 15.\n- 15, presence in the official MCP registry under a verified namespace (MCP servers), or current official SDKs (APIs and models).\n- 10, package health, current dependencies and CI.",
          "Models are read for deprecation notice periods and model churn rather than release counts."
        ],
        "checklistUrl": "https://www.anchorterminal.com/benchmark/#checklist-maintenance"
      }
    ],
    "unchecked": [
      "Whether Hobby signup needs a card. We didn't confirm it this run.",
      "Whether the Sandbox endpoints appear in Vercel's public OpenAPI file.",
      "Whether Sandbox calls show up in Vercel's audit log, and on which plans.",
      "What changed between @vercel/sandbox 2.7.0 (16 July 2026) and 3.x. The GitHub releases page we saw stopped at 2.7.0.",
      "The control-plane limits and the snapshot and inactivity retention periods come from last week's listing research on the pricing and persistent-sandboxes pages, which we didn't reload this run because of fetch limits."
    ],
    "weaknesses": [
      "Tied to a Vercel team and project even when called from elsewhere, and access tokens reach the whole team",
      "Hobby caps sessions at 45 minutes and pauses creation once the monthly allowance is spent",
      "Snapshots keep the filesystem, not memory or running processes",
      "Egress is allow-all until you set a policy",
      "No MCP server for Sandbox and no public OpenAPI for its endpoints found"
    ],
    "agentNotes": [
      "Call `sandbox.stop()` when the task is done. Memory bills until the session ends",
      "Use `Sandbox.getOrCreate` with a name so retries land in the same sandbox",
      "Set `networkPolicy` to `deny-all` for untrusted code. The default is allow-all",
      "Put API keys in credential brokering rules, not in the sandbox environment",
      "Pass `persistent: false` for one-off runs so no snapshot is stored or billed"
    ],
    "requests": [
      {
        "text": "Document 429 and Retry-After",
        "reviews": 1
      },
      {
        "text": "Publish a Sandbox SLA",
        "reviews": 1
      },
      {
        "text": "per-sandbox scoped tokens",
        "reviews": 1
      }
    ],
    "recheck": "https://www.anchorterminal.com/builders/#disputes"
  },
  "meta": {
    "attribution": "Anchor Terminal (https://www.anchorterminal.com)",
    "docs": "https://www.anchorterminal.com/docs/",
    "generatedAt": "2026-10-04",
    "license": "CC-BY-4.0",
    "method": "https://www.anchorterminal.com/benchmark/",
    "methodology": "0.3",
    "openapi": "https://www.anchorterminal.com/openapi.json",
    "preview": false,
    "run": "2026-10-01",
    "runLabel": "October 2026 research run"
  }
}