{
 "features_resolved": {
  "snapshot_any": false,
  "snapshot_mem": false,
  "fork": false,
  "pause_resume": true,
  "persistent_disk": true,
  "whole_disk": null,
  "persist_scope": null,
  "volumes": false,
  "auto_stop_idle": false,
  "long_sessions": true,
  "custom_image": null,
  "image_snapshot": false,
  "image_oci": false,
  "vm_isolation": null,
  "docker_inside": false,
  "nested_virt": false,
  "root": null,
  "systemd": null,
  "browser": true,
  "desktop": true,
  "computer_use": false,
  "code_interpreter": null,
  "browser_control": null,
  "desktop_control": null,
  "browser_and_desktop": false,
  "anti_bot": false,
  "captcha_solving": false,
  "residential_ip": null,
  "agent_harnesses": false,
  "ssh": false,
  "public_ipv4": false,
  "inbound_https": false,
  "custom_domain": null,
  "raw_tcp_inbound": null,
  "egress_allowlist": false,
  "open_internet": null,
  "static_egress_ip": false,
  "private_network": null,
  "self_host": false,
  "byoc": true,
  "open_source": false,
  "soc2": null,
  "hipaa": null,
  "sso": null,
  "gpu": false,
  "arm64": null,
  "windows": false,
  "macos": false,
  "lang_python": null,
  "lang_node": null,
  "lang_go": null,
  "lang_rust": null,
  "lang_java": null,
  "gpu_desktop": null,
  "credential_injection": null,
  "egress_http_rules": false,
  "wake_on_request": false,
  "live_resize": false,
  "fork_running": false,
  "memory_snapshot_fork": false,
  "webhooks": null,
  "scheduled_wakeups": true,
  "audit_logs": null,
  "eu_data_residency": null,
  "zero_data_retention": null,
  "scoped_api_keys": null,
  "spend_limits": true,
  "mcp_server": true,
  "usage_api": false,
  "snapshot_auto": false,
  "snapshot_on_demand": false,
  "harness_api": false,
  "own_agent_api": true,
  "fast_boot": null,
  "strong_isolation": null,
  "ingress_rules": false,
  "guest_firewall": null,
  "secret_proxy": false,
  "secret_proxy_any": false,
  "volume_attach": false,
  "volume_shared": false,
  "sidecar": null,
  "fuse": null,
  "kernel_features": null,
  "burst": null
 },
 "company": {
  "id": "dexto",
  "checked": "2026-09-30",
  "company": "Dexto (Truffle AI, Inc.)",
  "ownership": {
   "value": "independent",
   "parent": null,
   "source": "https://www.ycombinator.com/companies/dexto",
   "quote": "Dexto Founded: 2024 Batch: Winter 2025 Team Size: 2 Status: Active Location: San Francisco Primary Partner: Harj Taggar"
  },
  "stage": {
   "value": "pre-seed",
   "source": "https://www.ycombinator.com/companies/dexto",
   "quote": "Dexto: AWS for AI Agents | Y Combinator ... Batch: Winter 2025",
   "confidence": "low",
   "note": "No priced round is publicly disclosed. Company is a Winter 2025 (W25) Y Combinator batch company (team of 2, founded 2024 as a Truffle AI spin-off), which implies at least YC's standard pre-seed investment, but no amount, additional investors, or later round has been found in press, Crunchbase, or the company's own site."
  },
  "accelerators": [
   {
    "name": "Y Combinator",
    "batch": "Winter 2025",
    "source": "https://www.ycombinator.com/companies/dexto",
    "quote": "Dexto Founded: 2024 Batch: Winter 2025 Team Size: 2 Status: Active Location: San Francisco",
    "confidence": "high",
    "via": "research"
   }
  ],
  "total_funding_usd": {
   "value": null,
   "status": "unknown"
  },
  "last_round": {
   "type": "unknown",
   "amount_usd": null,
   "date": null,
   "leads": [],
   "status": "unknown",
   "source": "https://www.ycombinator.com/companies/dexto",
   "quote": "Primary Partner: Harj Taggar"
  },
  "investors": [
   "Y Combinator"
  ],
  "portfolio_listings": [],
  "revenue": {
   "value": null,
   "status": "unknown"
  },
  "employees": {
   "value": 2,
   "source": "https://www.ycombinator.com/companies/dexto"
  },
  "note": "Dexto is built by Truffle AI, Inc. (YC W25), founded by Shaunak Srivastava (CEO) and Rahul Karajgikar (CTO); Rahul was previously an AWS OpenSearch engineer. The company also maintains an open-source agent runtime (github.com/truffle-ai/dexto) separate from the hosted cloud-sandbox/agent-OS product reviewed here. Confirmed via YC's own company page that dexto.ai is the same company as the YC directory listing (website https://www.dexto.ai/ shown on the YC page). No funding amount, lead investor, or revenue figures are publicly available as of this check; extensive web/DuckDuckGo/Bing searches turned up no TechCrunch/press coverage of a funding round, and Crunchbase/Tracxn pages could not be retrieved (blocked)."
 },
 "id": "dexto",
 "name": "Dexto",
 "url": "https://www.dexto.ai/docs/models/",
 "category": "agent-sandbox",
 "isolation": null,
 "modes": [
  {
   "key": "native",
   "label": "Native product; see regime notes",
   "flags": [
    "alt"
   ],
   "os": [
    "linux"
   ],
   "regions": null,
   "gpu_only": false,
   "features": {},
   "pricing": "resource",
   "vcpu_h": null,
   "ram_gib_h": null,
   "cpu_basis": "alloc",
   "ram_basis": "alloc",
   "active_floor": 0,
   "ram_per_vcpu": [
    null,
    null
   ],
   "ram_fixed_per_vcpu": null,
   "vcpu_options": null,
   "min_vcpu": null,
   "max_vcpu": null,
   "max_ram_gib": null,
   "sizes": null,
   "pool_tiers": null,
   "requires_always_on": false,
   "always_on_month_per_instance": null,
   "commit_note": null,
   "min_billed_seconds": null,
   "granularity_s": null,
   "start_fee": null,
   "boot_overhead_s": null,
   "multiplier": 1,
   "windows_vcpu_h": null,
   "gpu": null,
   "note": "Incomplete or non-resource economics: rates remain null, never interpreted as free."
  }
 ],
 "storage": {
  "included_disk_gib": null,
  "disk_gib_month": null,
  "disk_billed_when_stopped": null,
  "snapshot_gib_month": null,
  "paused_ram_billed": null
 },
 "network": {
  "egress_free_gib": null,
  "egress_gib": null,
  "ipv4_month": null,
  "ipv4_available": null
 },
 "plans": [
  {
   "name": "Prepaid usage",
   "fee": 0,
   "flags": [],
   "fee_is_credit": false,
   "included_usd": 0,
   "per_seat": null,
   "concurrency": 1,
   "max_session_h": null,
   "max_vcpu": null,
   "max_ram_gib": null,
   "sizes_allowed": null,
   "note": ""
  },
  {
   "name": "Enterprise",
   "fee": null,
   "flags": [
    "sales"
   ],
   "fee_is_credit": false,
   "included_usd": null,
   "per_seat": null,
   "concurrency": null,
   "max_session_h": null,
   "max_vcpu": null,
   "max_ram_gib": null,
   "sizes_allowed": null,
   "note": ""
  }
 ],
 "free": {
  "monthly_credit": null,
  "one_time_credit": 5
 },
 "features": {
  "linux": true,
  "windows": false,
  "macos": false,
  "arm64": null,
  "gpu": false,
  "snapshot": "none",
  "pause_resume": true,
  "fork": false,
  "persistent_disk": true,
  "volumes": null,
  "auto_stop_idle": false,
  "max_session_h": null,
  "custom_image": null,
  "docker_inside": false,
  "nested_virt": false,
  "root": null,
  "systemd": null,
  "browser": true,
  "desktop": true,
  "computer_use": false,
  "code_interpreter": null,
  "agent_harnesses": false,
  "ssh": false,
  "public_ipv4": false,
  "inbound_https": false,
  "custom_domain": null,
  "raw_tcp_inbound": null,
  "egress_allowlist": false,
  "static_egress_ip": false,
  "private_network": null,
  "self_host": false,
  "byoc": true,
  "open_source": false,
  "soc2": null,
  "hipaa": null,
  "sso": null,
  "gpu_types": [],
  "regions": null,
  "lang_python": null,
  "lang_node": null,
  "lang_go": null,
  "lang_rust": null,
  "lang_java": null,
  "gpu_desktop": null,
  "credential_injection": null,
  "egress_http_rules": false,
  "wake_on_request": false,
  "live_resize": false,
  "fork_running": false,
  "memory_snapshot_fork": false,
  "webhooks": null,
  "scheduled_wakeups": true,
  "audit_logs": null,
  "eu_data_residency": null,
  "zero_data_retention": null,
  "scoped_api_keys": null,
  "spend_limits": true,
  "mcp_server": true,
  "own_agent_api": true,
  "image_oci": false,
  "anti_bot": false,
  "captcha_solving": false,
  "snapshot_auto": false,
  "snapshot_on_demand": false,
  "volume_attach": false,
  "volume_shared": false,
  "harness_api": false,
  "ingress_rules": false,
  "secret_proxy": false,
  "max_session_unlimited": true,
  "usage_api": false
 },
 "perf": {
  "cpu_runs_s": null,
  "cold_start_ms": null,
  "burst_tti_ms": null,
  "dax_s": null
 },
 "caveats": [
  "Rendered pricing links through /docs/pricing to Services billing: Sandbox managed compute $0.414/hour while running. CPU/RAM allocation not specified, so not normalized to 4/8.",
  "Public API permits one primary Linux sandbox per user; repeated create returns the same resource. Pause stops active compute charges while retaining workspace; visual desktop is not a separate public API resource.",
  "Free $5 one-time; Pro displayed $25 is top-up credit, not a monthly fee. Credits also fund tokens/media/connections; dedicated GPU/model services available on request.",
  "Null values mean unknown, not free/unlimited. No model-token cost, tax or unverified ancillary charge included.",
  "Native browser/device/task/credit prices with unknown CPU/RAM are deliberately non-priceable. Check regime before interpreting any partial resource estimate.",
  "Credits are prepaid and non-refundable. The Pro top-up is $25 by default, adjustable from $5 to $500. No expiry is stated on the public pages."
 ],
 "sources": [
  "https://www.ycombinator.com/companies/dexto",
  "https://www.dexto.ai/docs/models/",
  "https://www.dexto.ai/docs/platform/api/sandboxes/",
  "https://www.dexto.ai/pricing/",
  "https://www.dexto.ai/docs/",
  "https://www.dexto.ai/llms.txt"
 ],
 "fleet": {
  "default_max_instances": 1,
  "default_max_vcpu": null,
  "default_max_gpus": null,
  "raise": "none",
  "raise_time_days": null,
  "verification": null,
  "tiers": null,
  "notes": "Public CPU sandbox API gives each user one primary sandbox; repeat POST returns that same sandbox rather than provisioning another. No multi-sandbox creation/raise path documented; none denotes current API single-primary restriction, not a promise that private enterprise arrangements can never exist. Sandbox lifecycle API rate limits: create/start/pause/delete/processes 60 requests per 60 s (sandbox_write), reads 300 per 60 s; each command runs at most 120 s (timeout_ms max 120000).",
  "sources": [
   "https://www.dexto.ai/docs/platform/api/sandboxes/"
  ],
  "researched": "2026-09-28"
 },
 "audit_2026_09_29": {
  "applied": 2,
  "source": "checked against the provider's live pricing and docs pages",
  "reviewed": 6
 },
 "feature_evidence": {
  "own_agent_api": {
   "value": true,
   "source": "https://www.dexto.ai/skill.md",
   "quote": "Use `autobots.send`, `autobots.status`, and `autobots.read` for conversations. Call `autobots.respond` only for a pending input or approval returned for that owned AutoBot.",
   "note": "Only for Dexto's OWN agents (Dexto assistant / agent profiles / AutoBots), driven via the OAuth MCP server at https://app.dexto.ai/mcp ('Capabilities perform direct actions; Agents and AutoBots take on delegated work') or page-bound WebMCP tools on autobots.dexto.ai. The agent runs in the Dexto control plane and uses the cloud computer as its execution workspace, rather than being a harness process inside the sandbox. The REST API only lists agents (GET /api/agents); no REST endpoint to send prompts. No Claude Code/Codex/OpenCode harness driving. Separately, the open-source Dexto agent runtime (truffle-ai/dexto) has a self-hosted `dexto --mode server` 'REST & SSE APIs' mode, unrelated to the hosted sandbox.",
   "checked": "2026-09-29"
  },
  "image_oci": {
   "value": false,
   "source": "https://www.dexto.ai/docs/platform/api/sandboxes/",
   "quote": "Get or create your sandbox ... POST /v1/sandboxes ... The first call creates a paused sandbox and returns 201; later calls return the same primary sandbox with 200.",
   "note": "Sandbox creation takes no image parameter; each user gets one fixed primary Linux workspace, not a chosen/custom OCI image.",
   "checked": "2026-09-30"
  },
  "docker_inside": {
   "value": false,
   "source": "https://www.dexto.ai/docs/platform/api/sandboxes/",
   "quote": "POST /v1/sandboxes/{sandbox_id}/processes ... The response contains exit_code, stdout, and stderr.",
   "note": "not offered: outside the product's scope; the documented API only exposes plain command execution and file I/O under /workspace, with no mention of Docker/containers inside.",
   "checked": "2026-09-30"
  },
  "nested_virt": {
   "value": false,
   "source": "https://www.dexto.ai/docs/platform/api/sandboxes/",
   "quote": "A CPU sandbox is a Linux workspace you can start when you need to run code.",
   "note": "not offered: outside the product's scope; no /dev/kvm or nested virtualization capability is documented for the CPU sandbox.",
   "checked": "2026-09-30"
  },
  "computer_use": {
   "value": false,
   "source": "https://www.dexto.ai/docs/platform/api/sandboxes/",
   "quote": "Use your API key with https://api.dexto.ai; the dashboard's visual desktop is not a separate public API resource.",
   "note": "not offered: the visual desktop exists in the dashboard but there is no public screenshot/click API to drive it.",
   "checked": "2026-09-30"
  },
  "anti_bot": {
   "value": false,
   "source": "https://www.dexto.ai/docs/",
   "quote": "Dexto has its own persistent cloud computer that it uses to store files, run code, open browsers and",
   "note": "not offered: outside the product's scope; Dexto is an agent-work control plane, not a stealth/anti-detection browsing product.",
   "checked": "2026-09-30"
  },
  "captcha_solving": {
   "value": false,
   "source": "https://www.dexto.ai/docs/",
   "quote": "Dexto has its own persistent cloud computer that it uses to store files, run code, open browsers and",
   "note": "not offered: outside the product's scope.",
   "checked": "2026-09-30"
  },
  "ssh": {
   "value": false,
   "source": "https://www.dexto.ai/docs/platform/api/sandboxes/",
   "quote": "Use your API key with https://api.dexto.ai; the dashboard's visual desktop is not a separate public API resource.",
   "note": "not offered: the only documented access paths are the REST processes/files endpoints; no SSH endpoint or credentials are documented.",
   "checked": "2026-09-30"
  },
  "snapshot": {
   "value": "none",
   "source": "https://www.dexto.ai/docs/platform/api/sandboxes/",
   "quote": "Use POST /v1/sandboxes/{sandbox_id}/pause to stop charging for active compute while retaining its workspace.",
   "note": "Pause just stops billing while keeping the persistent workspace; no snapshot/checkpoint API (disk or memory) is documented.",
   "checked": "2026-09-30"
  },
  "snapshot_auto": {
   "value": false,
   "source": "https://www.dexto.ai/docs/platform/api/sandboxes/",
   "quote": "Use POST /v1/sandboxes/{sandbox_id}/pause to stop charging for active compute while retaining its workspace.",
   "note": "No auto-snapshot-on-stop mechanism documented beyond simply retaining the workspace files.",
   "checked": "2026-09-30"
  },
  "snapshot_on_demand": {
   "value": false,
   "source": "https://www.dexto.ai/docs/platform/api/sandboxes/",
   "quote": "See /v1/openapi.json for the complete request and response schemas, errors, and rate limits.",
   "note": "No snapshot/checkpoint endpoint is listed in the documented sandbox API surface.",
   "checked": "2026-09-30"
  },
  "fork": {
   "value": false,
   "source": "https://www.dexto.ai/docs/platform/api/sandboxes/",
   "quote": "The first call creates a paused sandbox and returns 201; later calls return the same primary sandbox with 200.",
   "note": "Each user has one primary sandbox and repeated create calls return that same resource; there is no clone/fork mechanism.",
   "checked": "2026-09-30"
  },
  "fork_running": {
   "value": false,
   "source": "https://www.dexto.ai/docs/platform/api/sandboxes/",
   "quote": "The first call creates a paused sandbox and returns 201; later calls return the same primary sandbox with 200.",
   "note": "Follows from no fork capability at all.",
   "checked": "2026-09-30"
  },
  "memory_snapshot_fork": {
   "value": false,
   "source": "https://www.dexto.ai/docs/platform/api/sandboxes/",
   "quote": "The first call creates a paused sandbox and returns 201; later calls return the same primary sandbox with 200.",
   "note": "Follows from no fork/snapshot capability at all.",
   "checked": "2026-09-30"
  },
  "wake_on_request": {
   "value": false,
   "source": "https://www.dexto.ai/docs/platform/api/sandboxes/",
   "quote": "POST /v1/sandboxes/{sandbox_id}/start",
   "note": "Starting a paused sandbox requires an explicit start call; no documentation of automatic wake on an inbound request.",
   "checked": "2026-09-30"
  },
  "volume_attach": {
   "value": false,
   "source": "https://www.dexto.ai/docs/platform/api/sandboxes/",
   "quote": "All file paths must stay under /workspace, and the sandbox must already be running.",
   "note": "Only a single fixed /workspace path is documented; no attachable extra volumes.",
   "checked": "2026-09-30"
  },
  "volume_shared": {
   "value": false,
   "source": "https://www.dexto.ai/docs/platform/api/sandboxes/",
   "quote": "Each user currently has one primary sandbox.",
   "note": "With one primary sandbox per user and no volume API, a shared multi-machine volume is not offered.",
   "checked": "2026-09-30"
  },
  "live_resize": {
   "value": false,
   "source": "https://www.dexto.ai/docs/models/",
   "quote": "Compute Sandbox ... Managed compute for agent runs. While running $0.414 / hour",
   "note": "No vCPU/RAM sizing options or resize endpoint are documented; pricing is a single flat hourly rate.",
   "checked": "2026-09-30"
  },
  "agent_harnesses": {
   "value": false,
   "source": "https://www.dexto.ai/docs/mcp/",
   "quote": "MCP setup: Connect Claude, ChatGPT/Codex, Cursor, OpenClaw, or Hermes through https://app.dexto.ai/mcp and OAuth.",
   "note": "not offered: outside the product's scope; Dexto connects to external coding harnesses via MCP rather than preinstalling harnesses like Claude Code/Codex/OpenCode inside its own sandbox.",
   "checked": "2026-09-30"
  },
  "harness_api": {
   "value": false,
   "source": "https://www.dexto.ai/docs/platform/api/agents/",
   "quote": "Agents: List API-key-visible agents.",
   "note": "not offered: the REST API only lists agents; there is no first-party API to prompt/stream/follow-up a coding harness running inside the sandbox.",
   "checked": "2026-09-30"
  },
  "public_ipv4": {
   "value": false,
   "source": "https://www.dexto.ai/docs/platform/api/sandboxes/",
   "quote": "Use your API key with https://api.dexto.ai; the dashboard's visual desktop is not a separate public API resource.",
   "note": "not offered: outside the product's scope; the sandbox is reached only through the Dexto API, with no per-machine public IPv4 documented.",
   "checked": "2026-09-30"
  },
  "inbound_https": {
   "value": false,
   "source": "https://www.dexto.ai/docs/platform/api/sandboxes/",
   "quote": "See /v1/openapi.json for the complete request and response schemas, errors, and rate limits.",
   "note": "No documented way to expose an HTTPS URL to a port on the sandbox; only processes/files endpoints exist.",
   "checked": "2026-09-30"
  },
  "ingress_rules": {
   "value": false,
   "source": "https://www.dexto.ai/docs/platform/api/sandboxes/",
   "quote": "See /v1/openapi.json for the complete request and response schemas, errors, and rate limits.",
   "note": "not offered: outside the product's scope; there is no exposed-port/ingress feature to control.",
   "checked": "2026-09-30"
  },
  "egress_http_rules": {
   "value": false,
   "source": "https://www.dexto.ai/docs/platform/api/sandboxes/",
   "quote": "A CPU sandbox is a Linux workspace you can start when you need to run code.",
   "note": "not offered: outside the product's scope; no method/path-level egress control is documented.",
   "checked": "2026-09-30"
  },
  "secret_proxy": {
   "value": false,
   "source": "https://www.dexto.ai/docs/platform/primitives/",
   "quote": "Credentials remain inside platform-controlled execution paths; an agent receives the permitted tool surface, not a",
   "note": "Connections keep third-party credentials inside Dexto's control plane for its own agents/capabilities, not a first-party HTTP proxy that injects secrets into arbitrary calls made from inside the CPU sandbox.",
   "checked": "2026-09-30"
  },
  "gpu": {
   "value": false,
   "source": "https://www.dexto.ai/docs/models/",
   "quote": "Compute Sandbox ... Managed compute for agent runs. While running $0.414 / hour. Available on request: Self-hosted models and dedicated GPU access for your team.",
   "note": "not offered: the public Compute Sandbox is CPU-only ('CPU sandboxes API'); dedicated GPU access is a separate enterprise model-hosting offering, not a GPU sandbox/VM product.",
   "checked": "2026-09-30"
  },
  "windows": {
   "value": false,
   "source": "https://www.dexto.ai/docs/platform/api/sandboxes/",
   "quote": "A CPU sandbox is a Linux workspace you can start when you need to run code.",
   "note": "not offered: outside the product's scope; the sandbox is explicitly a Linux workspace.",
   "checked": "2026-09-30"
  },
  "macos": {
   "value": false,
   "source": "https://www.dexto.ai/docs/platform/api/sandboxes/",
   "quote": "A CPU sandbox is a Linux workspace you can start when you need to run code.",
   "note": "not offered: outside the product's scope; the sandbox is explicitly a Linux workspace.",
   "checked": "2026-09-30"
  },
  "self_host": {
   "value": false,
   "source": "https://www.dexto.ai/docs/platform/",
   "quote": "Dexto Cloud routes LLMs, image generation, voice and audio, video, open-source LLMs, and managed compute through one services catalog.",
   "note": "not offered: outside the product's scope; the hosted Dexto Cloud sandbox/control plane has no documented self-hosting option (a separate, unrelated open-source truffle-ai/dexto agent runtime exists but is not this hosted sandbox product).",
   "checked": "2026-09-30"
  },
  "open_source": {
   "value": false,
   "source": "https://www.dexto.ai/docs/platform/",
   "quote": "Dexto Cloud routes LLMs, image generation, voice and audio, video, open-source LLMs, and managed compute through one services catalog.",
   "note": "The hosted Dexto Cloud control plane / sandbox product itself is not published as open source; only a separate, unrelated agent-runtime repo (truffle-ai/dexto) is open source.",
   "checked": "2026-09-30"
  },
  "auto_stop_idle": {
   "value": false,
   "source": "https://www.dexto.ai/docs/platform/api/sandboxes/",
   "quote": "Use POST /v1/sandboxes/{sandbox_id}/pause to stop charging for active compute while retaining its workspace.",
   "note": "Only a manual, explicit pause call is documented for stopping active-compute billing; no idle-detection or automatic-stop behavior is mentioned anywhere in the sandbox lifecycle docs or OpenAPI schema.",
   "checked": "2026-09-30"
  },
  "max_session_h": {
   "value": "none",
   "source": "https://www.dexto.ai/docs/platform/api/sandboxes/",
   "quote": "Use POST /v1/sandboxes/{sandbox_id}/pause to stop charging for active compute while retaining its workspace.",
   "note": "No maximum running duration is documented anywhere; the only lifecycle controls are manual start/pause/delete (per-command execution is capped at timeout_ms<=120000, i.e. 2 minutes per command, but that's a command timeout, not a sandbox session limit).",
   "checked": "2026-09-30"
  },
  "egress_allowlist": {
   "value": false,
   "source": "https://www.dexto.ai/docs/platform/api/sandboxes/",
   "quote": "See /v1/openapi.json for the complete request and response schemas, errors, and rate limits.",
   "note": "not offered: outside the product's scope as documented; the CPU sandbox API surface is limited to sandbox lifecycle, command execution, and file operations under /workspace, with no egress/allowlist/security-group configuration endpoint or docs mentioned anywhere.",
   "checked": "2026-09-30"
  },
  "static_egress_ip": {
   "value": false,
   "source": "https://www.dexto.ai/docs/platform/api/sandboxes/",
   "quote": "See /v1/openapi.json for the complete request and response schemas, errors, and rate limits.",
   "note": "not offered: outside the product's scope as documented; no network configuration (static IP, egress IP, etc.) is exposed anywhere in the sandbox API or docs.",
   "checked": "2026-09-30"
  },
  "usage_api": {
   "value": false,
   "source": "https://api.dexto.ai/v1/openapi.json",
   "quote": "Returns the caller organization's current credit balance, charged usage for the preceding 30 days, and up to 50 recent model usage events.",
   "note": "granularity: account-only; cost: yes; GET /v1/billing/summary endpoint returns account-level billing with recent_model_usage events including cost_usd, input_tokens, output_tokens, model, and timestamp. No per-sandbox, per-tag, or per-project breakdown is available; granularity is organization-wide.",
   "confidence": "high",
   "checked": "2026-10-01"
  }
 },
 "locations": [],
 "locations_source": "https://www.dexto.ai/docs/platform/api/sandboxes/",
 "locations_quote": "A CPU sandbox is a Linux workspace you can start when you need to run code. Each user currently has one primary sandbox. Use your API key with https://api.dexto.ai; the dashboard's visual desktop is not a separate public API resource.",
 "locations_checked": "2026-10-01"
}