← AgentiSendCONTENT HISTORY

Update to AgentiSend

Snapshot Oct 31, 2026 · 06:00 UTC · version 1.1.0

WHAT CHANGED

Declared capabilities changed

Full description — Now says: API for AI agents: send receipts, password resets, sign-in codes and alerts from your own verified domain, with a message log, a preflight that checks a send...

Observed in instructions or declared skills. Runtime behavior has not been tested.

Save AgentiSend and check future changes in Following.

Up to 10 plugins free · no card required.

All fields and original values.

Full description

Before

Oct 24, 2026 · 12:40 UTC

API: send through verified domains with a message log, a free preflight that runs every gate a real send runs, and refusals that name the fix. Built so an agent can send on your behalf without ever exceeding a budget a person set. Unsoli...

After

Oct 31, 2026 · 06:00 UTC

API for AI agents: send receipts, password resets, sign-in codes and alerts from your own verified domain, with a message log, a preflight that checks a send without sending it, sending limits per key, an approval queue and refusals that...

Long description

Before

Oct 24, 2026 · 12:40 UTC

API: send through verified domains with a message log, a free preflight that runs every gate a real send runs, and refusals that name the fix. Built so an agent can send on your behalf without ever exceeding a budget a person set. Unsoli...

After

Oct 31, 2026 · 06:00 UTC

API for AI agents: send receipts, password resets, sign-in codes and alerts from your own verified domain, with a message log, a preflight that runs every check a real send runs without sending anything, and refusals that name the fix. B...

Declared skills

Before

Oct 24, 2026 · 12:40 UTC

[]

After

Oct 31, 2026 · 06:00 UTC

[{"description":"AgentiSend is a transactional email API for AI agents. Use when asked to send a receipt, password reset, sign-in code, alert or any transactional email, to check what an email send would do before doing it, to find out w...

Agentisend

Before

Oct 24, 2026 · 12:40 UTC

Not present

After

Oct 31, 2026 · 06:00 UTC

{"id":"asdk_app_6aad8b5cad4c8191b02f399b0cdf2575","required":true}

Compare saved observations

Download comparison JSON
Full technical diff · 17 changed fields

added /release/app_manifest/apps/agentisend

BEFORE
Field was absent
AFTER
{
  "id": "asdk_app_6aad8b5cad4c8191b02f399b0cdf2575",
  "required": true
}

removed /release/app_manifest/apps/app-6aad8b5cad4c8191b02f399b0cdf2575

BEFORE
{
  "id": "asdk_app_6aad8b5cad4c8191b02f399b0cdf2575",
  "required": true
}
AFTER
Field is absent

changed /release/description

BEFORE
"AgentiSend is a transactional email API: send through verified domains with a message log, a free preflight that runs every gate a real send runs, and refusals that name the fix. Built so an agent can send on your behalf without ever exceeding a budget a person set. Unsolicited mail is refused by design; there is no tool for it."
AFTER
"AgentiSend is a transactional email API for AI agents: send receipts, password resets, sign-in codes and alerts from your own verified domain, with a message log, a preflight that checks a send without sending it, sending limits per key, an approval queue and refusals that name the fix."

changed /release/id

BEFORE
"pluginrel_bf7b0bcafd2081919245d0ba27f82560"
AFTER
"pluginrel_f4771d10a3b08191abe0adce586213d7"

changed /release/interface/capabilities

BEFORE
[]
AFTER
[
  "Send a receipt, password reset, sign-in code or alert from your verified domain",
  "Send up to 500 different emails in one call, with a result per item",
  "Schedule an email for later, then cancel or reschedule it",
  "Preflight a send: every check a real send runs, nothing sent",
  "Check how many emails this key may still send and when the limit resets",
  "Ask a person to approve a send before it goes out",
  "Find out why an email was not sent, and what fixes it",
  "Explain a bounce or complaint with the remediation as callable steps",
  "Read a message and its delivery timeline",
  "Add a sending domain and get the DNS records to publish",
  "Re-check DNS and wait for a domain to verify",
  "Honour unsubscribes: list, add and remove suppressed addresses",
  "Render and publish email templates",
  "Keep contacts and segments up to date",
  "Create and remove webhooks for delivery events",
  "Read the account trust standing and whether the kill switch is on"
]

changed /release/interface/composer_icon_dark_url

BEFORE
"https://files.openai.com/content?id=file_00000000c22c8208b3022cdcdb06aaaa"
AFTER
"https://files.openai.com/content?id=file_0000000027608208939c9100831cb1f9"

changed /release/interface/composer_icon_url

BEFORE
"https://files.openai.com/content?id=file_00000000c22c8208b3022cdcdb06aaaa"
AFTER
"https://files.openai.com/content?id=file_00000000e1188208b1bb7824eab458ca"

changed /release/interface/default_prompt

BEFORE
"Send the order receipt for order 4821 to reviewer@agentisend.com and tell me its status."
AFTER
"Send this receipt to the customer."

changed /release/interface/default_prompts

BEFORE
[
  "Send the order receipt for order 4821 to reviewer@agentisend.com and tell me its status.",
  "Run a preflight on a password-reset email to reviewer@agentisend.com; don't send.",
  "Why did the bounced message in this account bounce, and what fixes it?"
]
AFTER
[
  "Send this receipt to the customer.",
  "Preflight this send and tell me what would stop it.",
  "How many emails can this key still send today?"
]

changed /release/interface/developer_name

BEFORE
"WarmySender"
AFTER
"Agentisend FZCO"

changed /release/interface/logo_url

BEFORE
"https://files.openai.com/content?id=file_00000000a9e08208bba8a819aa85a495"
AFTER
"https://files.openai.com/content?id=file_00000000dba88211805f3e91457b3804"

changed /release/interface/logo_url_dark

BEFORE
"https://files.openai.com/content?id=file_00000000a9e08208bba8a819aa85a495"
AFTER
"https://files.openai.com/content?id=file_000000001ac482088f662fd566df43dc"

changed /release/interface/long_description

BEFORE
"AgentiSend is a transactional email API: send through verified domains with a message log, a free preflight that runs every gate a real send runs, and refusals that name the fix. Built so an agent can send on your behalf without ever exceeding a budget a person set. Unsolicited mail is refused by design; there is no tool for it."
AFTER
"AgentiSend is a transactional email API for AI agents: send receipts, password resets, sign-in codes and alerts from your own verified domain, with a message log, a preflight that runs every check a real send runs without sending anything, and refusals that name the fix. Built so an agent can send on your behalf inside a sending limit a person set: a loop guard holds repeated sends for a person, an approval queue lets you ask before a risky send, and a kill switch stops a key outright. Unsolicited mail is refused by design; there is no tool for it. Connect with OAuth, then say what to send."

changed /release/keywords

BEFORE
[]
AFTER
[
  "MCP server",
  "MCP connector",
  "transactional email API",
  "transactional email",
  "AI agents",
  "email API",
  "Claude",
  "Cursor",
  "sending domain",
  "email deliverability",
  "email for AI agents",
  "ChatGPT",
  "send email"
]

changed /release/onboarding_skill_name

BEFORE
null
AFTER
"agentisend"

changed /release/skills

BEFORE
[]
AFTER
[
  {
    "description": "AgentiSend is a transactional email API for AI agents. Use when asked to send a receipt, password reset, sign-in code, alert or any transactional email, to check what an email send would do before doing it, to find out why an email was not sent or bounced, to set up or verify a sending domain, or to ask a person before a risky send.",
    "interface": {
      "brand_color": null,
      "default_prompt": null,
      "display_name": "agentisend",
      "icon_large_url": null,
      "icon_small_url": null,
      "iconography": "chat",
      "short_description": "AgentiSend is a transactional email API for AI agents. Use when asked to send a receipt, password reset, sign-in code, alert or any transactional email, to check what an email send would do before doing it, to find out why an email was not sent or bounced, to set up or verify a sending domain, or to ask a person before a risky send."
    },
    "name": "agentisend",
    "plugin_release_skill_id": "pluginrsk_6ac11f7f69608191b01991248c54fc1a"
  }
]

changed /release/version

BEFORE
"1.0.0"
AFTER
"1.1.0"
Full snapshot data
{
  "canonical_app_id": "asdk_app_6aad8b5cad4c8191b02f399b0cdf2575",
  "connector_id": "asdk_app_6aad8b5cad4c8191b02f399b0cdf2575",
  "created_at": "2026-09-18T19:05:01.684092Z",
  "discoverability": "LISTED",
  "id": "plugin_asdk_app_6aad8b5cad4c8191b02f399b0cdf2575",
  "is_template": false,
  "is_template_instance": false,
  "name": "app-6aad8b5cad4c8191b02f399b0cdf2575",
  "release": {
    "app_ids": [
      "asdk_app_6aad8b5cad4c8191b02f399b0cdf2575"
    ],
    "app_manifest": {
      "apps": {
        "agentisend": {
          "id": "asdk_app_6aad8b5cad4c8191b02f399b0cdf2575",
          "required": true
        }
      }
    },
    "app_templates": [],
    "description": "AgentiSend is a transactional email API for AI agents: send receipts, password resets, sign-in codes and alerts from your own verified domain, with a message log, a preflight that checks a send without sending it, sending limits per key, an approval queue and refusals that name the fix.",
    "display_name": "AgentiSend",
    "id": "pluginrel_f4771d10a3b08191abe0adce586213d7",
    "interface": {
      "brand_color": null,
      "capabilities": [
        "Send a receipt, password reset, sign-in code or alert from your verified domain",
        "Send up to 500 different emails in one call, with a result per item",
        "Schedule an email for later, then cancel or reschedule it",
        "Preflight a send: every check a real send runs, nothing sent",
        "Check how many emails this key may still send and when the limit resets",
        "Ask a person to approve a send before it goes out",
        "Find out why an email was not sent, and what fixes it",
        "Explain a bounce or complaint with the remediation as callable steps",
        "Read a message and its delivery timeline",
        "Add a sending domain and get the DNS records to publish",
        "Re-check DNS and wait for a domain to verify",
        "Honour unsubscribes: list, add and remove suppressed addresses",
        "Render and publish email templates",
        "Keep contacts and segments up to date",
        "Create and remove webhooks for delivery events",
        "Read the account trust standing and whether the kill switch is on"
      ],
      "category": "Developer Tools",
      "composer_icon_dark_url": "https://files.openai.com/content?id=file_0000000027608208939c9100831cb1f9",
      "composer_icon_url": "https://files.openai.com/content?id=file_00000000e1188208b1bb7824eab458ca",
      "default_prompt": "Send this receipt to the customer.",
      "default_prompts": [
        "Send this receipt to the customer.",
        "Preflight this send and tell me what would stop it.",
        "How many emails can this key still send today?"
      ],
      "developer_name": "Agentisend FZCO",
      "logo_url": "https://files.openai.com/content?id=file_00000000dba88211805f3e91457b3804",
      "logo_url_dark": "https://files.openai.com/content?id=file_000000001ac482088f662fd566df43dc",
      "long_description": "AgentiSend is a transactional email API for AI agents: send receipts, password resets, sign-in codes and alerts from your own verified domain, with a message log, a preflight that runs every check a real send runs without sending anything, and refusals that name the fix. Built so an agent can send on your behalf inside a sending limit a person set: a loop guard holds repeated sends for a person, an approval queue lets you ask before a risky send, and a kill switch stops a key outright. Unsolicited mail is refused by design; there is no tool for it. Connect with OAuth, then say what to send.",
      "plugin_category_id": "developer tools",
      "privacy_policy_url": "https://agentisend.com/privacy",
      "screenshot_urls": [],
      "short_description": "Transactional email API",
      "terms_of_service_url": "https://agentisend.com/terms",
      "website_url": "https://agentisend.com"
    },
    "keywords": [
      "MCP server",
      "MCP connector",
      "transactional email API",
      "transactional email",
      "AI agents",
      "email API",
      "Claude",
      "Cursor",
      "sending domain",
      "email deliverability",
      "email for AI agents",
      "ChatGPT",
      "send email"
    ],
    "onboarding_skill_name": "agentisend",
    "requires_local_executor": false,
    "skills": [
      {
        "description": "AgentiSend is a transactional email API for AI agents. Use when asked to send a receipt, password reset, sign-in code, alert or any transactional email, to check what an email send would do before doing it, to find out why an email was not sent or bounced, to set up or verify a sending domain, or to ask a person before a risky send.",
        "interface": {
          "brand_color": null,
          "default_prompt": null,
          "display_name": "agentisend",
          "icon_large_url": null,
          "icon_small_url": null,
          "iconography": "chat",
          "short_description": "AgentiSend is a transactional email API for AI agents. Use when asked to send a receipt, password reset, sign-in code, alert or any transactional email, to check what an email send would do before doing it, to find out why an email was not sent or bounced, to set up or verify a sending domain, or to ask a person before a risky send."
        },
        "name": "agentisend",
        "plugin_release_skill_id": "pluginrsk_6ac11f7f69608191b01991248c54fc1a"
      }
    ],
    "version": "1.1.0"
  },
  "scope": "GLOBAL",
  "status": "ENABLED"
}

SHA-256 of public snapshot: dffb395d98adc126ce93bf2d7b8439163ba4dc2b4775cee6c56f44980a9863ec