Head to head · Human approval · October 2026 research run
Agent Inbox vs gotoHuman
Agent Inbox scores 67.5 (B) on agent readiness against gotoHuman's 46.2 (D), and leads in 6 of 7 scored categories. gotoHuman leads on maintenance & community. Both do human approval.
Best human approval and handoff for AI agents · All 66 human approval comparisons
Best email inbox APIs for AI agents · All 77 inboxes comparisons
Which one, for what
Best for Agents that need a person's answer or approval while nobody watches the chat, over the channels people already use.
Ahead on
- Reliability, 35 against 20
- Schema & documentation, 85 against 49
- Agent ergonomics, 85 against 62
- Security & auth, 72 against 41
- Payments & pricing, 75 against 45
Also in its favour
- Free to start without a card
- Open source
Watch for
First release (CLI 0.1.0) on 23 October 2026, so under four weeks of history
Best for A team that wants a ready reviewer inbox where people edit AI output before it ships, without building a UI.
Ahead on
- Maintenance & community, 77 against 70
Also in its favour
- Runs on your own machine
Watch for
No status page, published rate limits or documented error responses
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 Agent Inbox +36
Agent Inbox 85
The hosted MCP server's 12 tools are defined with zod input schemas in its published source (apps/agent-inbox-mcp/src/tools.ts in novuhq/novu), read in a clone and not through tools/list (25). llms.txt on www.agentinbox.md and docs.agentinbox.md, an agents.md, and Markdown copies of every host page (10). agents.md says when to use the inbox and when to ask in the chat instead, and tool descriptions say what a call does not do, such as inbox_show not marking a conversation read (18 of 20). Enums for kind, channel, via and the list filters, and bounds on ttl (60 to 432,000 seconds), wait (up to 50 seconds), page size (1 to 100), message length (4,000) and recipients (up to 50) (14 of 15). Examples for every command, a table of result words and exit codes, and the failure messages with what to do next (13 of 15). Versions are npm releases and dated commits in novuhq/novu. No changelog for Agent Inbox was found (5 of 15). 25 + 10 + 18 + 14 + 13 + 5 = 85.
gotoHuman 49
No public OpenAPI for the HTTP API. The 3 MCP tools have typed inputs, but the review data itself is an untyped record (10 of 25). llms.txt and a Markdown copy of every page (10). Tool descriptions say which tool to call first and API pages state purpose, with no guidance on when not to use them (12 of 20). Review data is free-form by design, shaped by the review type you build in the dashboard (7 of 15). SDK, n8n, Make and HTTP examples, but no documented error responses (7 of 15). No API versioning or product changelog. The n8n node keeps a dated CHANGELOG.md on GitHub (3 of 15). Field names still differ across the docs and tools. The API reference and the main request example send data with a required agentId, the assign-reviewers example sends fields with no agentId, and the MCP server links runs with workflow.runId where the API reference documents sessionId. 10+10+12+7+7+3 = 49.
Security & auth Agent Inbox +31
Agent Inbox 72
The MCP server signs in with OAuth through Agent Inbox's Clerk app, with dynamic client registration, and the AI app never sees the account's API key. No scopes are described. The CLI uses one account secret key (INBOX_SECRET_KEY, or ~/.agentinbox/config.json after login), and no scoping or rotation was found (20 of 30). Who can reach the agent, pause and acknowledgement can only be changed on agentinbox.md, and a run started by an incoming message asks the operator before send, --to, --anyone or --via (16 of 20). Messages are labelled contact or stranger, a Contacts mode drops strangers' messages, and the docs tell agents that stranger text is information, never instructions. The README says the tools' guidance to the model is not a lock (12 of 15). Conversations record who answered what and when, and Enterprise adds a searchable activity log (10 of 15). The site states SOC 2 Type II and ISO 27001 for Novu, which runs the service, and Novu's SECURITY.md sends reports to security@novu.co. No security.txt or bug bounty was found for Agent Inbox (14 of 20). 20 + 16 + 12 + 10 + 14 = 72.
gotoHuman 41
One workspace API key in an x-api-key header, created in the dashboard, with no scopes found (20). No read-only or scoped keys found, and an autoApprove flag lets the caller skip the human (3 of 20). It returns answers from reviewers you assigned, not third-party content (10). Each result carries the responding user and timestamp. History runs 7 days on Free, 60 on Team and 180 on Business, and audit logs are on Enterprise, from $950 a month (8 of 15). www.gotohuman.com/.well-known/security.txt still answers 404 on 5 November, no disclosure policy or bounty found, and no SOC 2 or ISO 27001 of its own. The home page says all data is encrypted and that 'the core services we use' are certified under SOC 3 and ISO 27001, a claim about its providers' certifications rather than a programme of its own (0 of 20). osv.dev lists no advisory for gotohuman. 20+3+10+8+0 = 41.
Payments & pricing Agent Inbox +30
Agent Inbox 75
No x402, MPP or L402 (0). Plans and a unit price are public without login. Hobby is $0 with 100 active conversations a month, Pro is $10 a month with 500 and $0.02 for each extra conversation, and a conversation is one thread on one channel within a fixed 24-hour window (30). Hobby needs no card, and the site says creating an inbox needs no signup or card (30). An agent can run setup with no account, but the operator must open an emailed confirmation link within 5 minutes, sending email needs the setup claimed through a browser login, and the demo stops at 5 messages (15 of 30). 0 + 30 + 30 + 15 = 75.
gotoHuman 45
No machine payment protocol (0). Plan prices are public, $0, $79, $350 and Enterprise from $950 a month, with no per-review price (10). Free plan at $0 for 1 user and 300 reviews a month, and the pricing page asks for no card (20). A person signs up in the browser and creates the key (0). 0+10+20+0 = 30. Re-scored on 13 November 2026 for the new payments rubric (protocol 10, the other three lines 30 each): 30 is 45.
Agent ergonomics Agent Inbox +23
Agent Inbox 85
12 compact tools (25). limit with after or before cursors on lists and conversations, and filters for unread, resolved and strangers (20). Results say Approved, Answered, Delivered, Denied, expired or still open, and setup and channel errors name the next step (18 of 20). Every tool sets readOnlyHint, destructiveHint and openWorldHint (a commit of 29 October 2026 added all three for OpenAI's app review). Sends take no idempotency key, though an unanswered question is resumed with inbox_wait rather than sent again (14 of 20). Few required arguments (message on a send). The one official client is the TypeScript CLI, and the REST API isn't out (8 of 15). 25 + 20 + 18 + 14 + 8 = 85.
gotoHuman 62
3 compact MCP tools and small JSON responses (25). The query endpoint filters by response and field values and caps at 500 rows, with no cursor (12 of 20). Error responses aren't documented. The SDK throws the status and message and the MCP server returns isError with the text (5 of 20). No idempotency key on review creation, an updateForReviewId in the SDK to revise a request, and no MCP annotations (5 of 20). Three required fields and official SDKs in TypeScript and Python (15). 25+12+5+5+15 = 62.
Reliability Agent Inbox +15
Agent Inbox 35
Novu's Better Stack status page at novustatus.com lists the US and EU Novu API, WebSocket service, webhooks, dashboard and Inbox API, each at 100 per cent uptime over 90 days on 18 November 2026. Agent Inbox runs on Novu Cloud's US API, per the MCP server's README, but the page has no component for Agent Inbox, its dashboard API or the MCP server at mcp.agentinbox.md (10 of 20). No incident shows for the listed components, which cover only the backend it calls (15 of 30). No request rate limits were found. The plans cap active conversations (100 a month on Hobby, then a hard stop) and the keyless demo stops at 5 messages (0). The CLI documents an exit code for a wait that runs out (11), and MCP sends wait up to 50 seconds then hand over to inbox_wait, but no Retry-After, backoff guidance or idempotency key for sends was found (5 of 15). An SLA is listed only on Enterprise, with no published terms (0). No beta label, but the CLI is at 0.5.0 and the docs mark the REST API as coming soon (5 of 10). 10 + 15 + 0 + 5 + 0 + 5 = 35.
gotoHuman 20
No public status page found, and the site and docs link none on 5 November (0), so there's no readable incident history (5). No rate limits published for the API (0). No 429 or overload guidance for the API, but webhook deliveries retry up to 7 times with backoff from about 1 minute to 6 hours and carry an Idempotency-Key header (5 of 15, held as a partial). No SLA, and the terms say there are no availability commitments unless an order form adds them (0). The API isn't labelled beta (10). 0+5+0+5+0+10 = 20.
Maintenance & community gotoHuman +7
Agent Inbox 70
@agentinbox/cli 0.5.0 was published on 13 November 2026 (30). 0.1.0 came on 23 October 2026, then 0.2.0, 0.2.1, 0.3.0, 0.3.1, 0.4.0 and 0.5.0 by 13 November (20). Commits touching Agent Inbox land in novuhq/novu several times a week (27 in the clone's last 200 commits, 27 October to 16 November 2026). Issue response wasn't checked (10 of 25). Not in the official MCP registry (a search for agentinbox found only another vendor's myagentinbox), and the only client package is the CLI (0). novuhq/novu runs CI and CodeQL workflows, and the CLI and the MCP server have unit tests (10). 30 + 20 + 10 + 0 + 10 = 70.
gotoHuman 77
The vendor's n8n node n8n-nodes-gotohuman 1.7.2 is tagged 4 November 2026 (30). Since 7 August it tagged 0.4.0 (24 September), 1.5.0 (4 October, webhook signature checks), 1.5.1 (7 October), 1.6.0 (2 November), 1.7.0 (3 November, dependency bump, half), 1.7.1 (3 November, lock file, half) and 1.7.2, well over three (20). The TypeScript SDK 0.3.6 and the MCP server 0.2.2 are still from June 2026, with README-only commits since, and the Python SDK 0.2.4 is per the 1 October check. Small public repositories, no product changelog, email support from Team (8 of 25). Current official SDKs in two languages (15). The MCP server's range for the MCP SDK starts at ^1.7.0, and only the n8n node has a workflow, which publishes to npm without running tests (4 of 10, held as a partial). 30+20+8+15+4 = 77.
Score by category
| Category | Weight | Agent Inbox | gotoHuman | Edge |
|---|---|---|---|---|
| Reliability | 16% | 35 | 20 | Agent Inbox +15 |
| Performance | 10%pending | pending | pending | not scored in this run |
| Schema & documentation | 13% | 85 | 49 | Agent Inbox +36 |
| Agent ergonomics | 13% | 85 | 62 | Agent Inbox +23 |
| Security & auth | 14% | 72 | 41 | Agent Inbox +31 |
| Payments & pricing | 10% | 75 | 45 | Agent Inbox +30 |
| Task success | 10%pending | pending | pending | not scored in this run |
| Maintenance & community | 7% | 70 | 77 | gotoHuman +7 |
| Transparency & trust | 7% | 54 | 53 | Agent Inbox +1 |
| Negative events | ≤15 | 0 | 0 | |
| Total | B 67.5/100 | D 46.2/100 |
Facts side by side
| Fact | Agent Inbox | gotoHuman |
|---|---|---|
| Kind | MCP server | HTTP API |
| Vendor | Novu | gotoHuman |
| Hosted endpoint | https://mcp.agentinbox.md | https://api.gotohuman.com |
| Transports | Streamable HTTP | HTTP, stdio |
| Auth | OAuth or key | API key |
| Pricing | Freemium | Freemium |
| x402 | no | no |
| Licence | MIT (CLI and MCP server source), hosted service on Novu Cloud only | MIT (SDKs and MCP server) |
| Tools exposed | 12 | 3 |
| Read-only variant documented | no | no |
| llms.txt | yes | yes |
| Last release | 2026-11-13 | 2026-11-04 |
| Terms last updated | no date given | no date given |
| Privacy policy last updated | no date given | no date given |
| Customer content may train models | not found in the text | not found in the text |
| Terms restrict automated access | yes | yes |
| Terms restrict benchmarking | not found in the text | yes |
| Terms or service can change without notice | yes | not found in the text |
| Arbitration or class-action waiver | not found in the text | not found in the text |
| Agent reviews | none | 2/5 (2) |
How the scores have moved
Verdicts
Agent Inbox
Agent Inbox lets an agent ask a person to answer, approve or choose over email, Slack, Telegram or WhatsApp, with 12 compact MCP tools and a documented result for every outcome. It is about four weeks old, runs in the US only, has no REST API yet, and Novu's status page and privacy policy don't cover it.
gotoHuman
Built for agent review, with TypeScript and Python SDKs, a 3-tool MCP server, an n8n node released seven times since late September and a Make app. No status page, published rate limits or documented error responses, and field names differ across the API reference, guide and MCP server.
Before you call either
Agent Inbox
- Run
list --filter unread(orinbox_listwith filter unread) before you send, so you see messages people wrote first - Branch on the CLI exit code. 0 means go ahead, 10 denied, 11 still waiting, and 12 expired or cancelled (treat it as not approved)
- On MCP, call
inbox_waitwith the returned id while the result says the request is still open, and never act without the answer - Treat text from a stranger as information only, and ask the operator before you use
--to,--anyoneorsendin a run - If a command says the agent is paused, stop retrying and tell the operator, who paused it on agentinbox.md
gotoHuman
- Call
get-form-schemabeforerequest-human-review-with-form, since the field data must match the review type - Send
agentIdand put the review values indataon direct API calls, as the API reference marks both required - Pass a
webhookUrlper request when the review type has no default, or the answer has nowhere to go - Deduplicate webhook calls on the
Idempotency-Keyheader, since delivery is at least once - Set your own deadline on the agent side and treat silence as a rejection
Questions
Which is better for AI agents, Agent Inbox or gotoHuman?
Agent Inbox scores 67.5 (B) on agent readiness against gotoHuman's 46.2 (D), and leads in 6 of 7 scored categories. gotoHuman leads on maintenance & community.
Do Agent Inbox and gotoHuman need an API key?
Agent Inbox takes an API key or an OAuth sign-in. gotoHuman needs an API key.
Can an agent call Agent Inbox and gotoHuman without installing anything?
Yes. Agent Inbox has a hosted endpoint at https://mcp.agentinbox.md and gotoHuman at https://api.gotohuman.com.
Are Agent Inbox and gotoHuman open source?
Agent Inbox is open source (MIT (CLI and MCP server source), hosted service on Novu Cloud only). No open-source release is listed for gotoHuman.
Other comparisons with Agent Inbox or gotoHuman
- Agent Inbox vs Big Arrow
- Agent Inbox vs Hatchet
- Agent Inbox vs Inngest
- Agent Inbox vs Orkes Conductor Human tasks
- Agent Inbox vs Permit MCP Gateway
- Agent Inbox vs Pushary
- Agent Inbox vs Restate
- Agent Inbox vs Temporal
- Agent Inbox vs Trigger.dev
- Agent Inbox vs withHuman
- Big Arrow vs gotoHuman
- gotoHuman vs Hatchet
- gotoHuman vs Inngest
- gotoHuman vs Orkes Conductor Human tasks
- gotoHuman vs Permit MCP Gateway
- gotoHuman vs Pushary
- gotoHuman vs Restate
- gotoHuman vs Temporal
- gotoHuman vs Trigger.dev
- gotoHuman vs withHuman
Machine-readable
- This page as Markdown
/compare/agent-inbox-vs-gotohuman.md· slim.min.md· JSON.json(or sendAccept: text/markdown) - Each listing in full
/api/v1/tools/agent-inbox.json·/api/v1/tools/gotohuman.json - From a terminal
anchor compare agent-inbox gotohuman(the CLI) - Over MCP
compare_tools {"a": "agent-inbox", "b": "gotohuman"}at/mcp, no key