dweet.io

{
  "service_type": "platform",
  "base_url": "https://dweet.io",
  "auth_method": "none",
  "auth_config": {
    "note": "Public dweets are readable/writable without authentication. Locked things and alerts require a lock/access key returned at creation time. Optional security via /lock and /unlock endpoints."
  },
  "endpoints": [
    {
      "name": "Dweet for a thing",
      "method": "POST",
      "path": "/dweet/for/{thing}",
      "description": "Create a new dweet (message) for the given thing name, with arbitrary key/value data."
    },
    {
      "name": "Get latest dweets for a thing",
      "method": "GET",
      "path": "/get/latest/dweet/for/{thing}",
      "description": "Retrieve the most recent dweet for a thing."
    },
    {
      "name": "Get all dweets for a thing",
      "method": "GET",
      "path": "/get/dweets/for/{thing}",
      "description": "Retrieve the most recent set of dweets for a thing."
    },
    {
      "name": "Read all dweets by thing",
      "method": "GET",
      "path": "/get/dweets/for/{thing}",
      "description": "Same as above; historical read access to a thing's stream."
    },
    {
      "name": "Lock a thing",
      "method": "GET",
      "path": "/lock/{thing}",
      "description": "Lock a thing to prevent others from dweeting to it; returns a lock key."
    },
    {
      "name": "Unlock a thing",
      "method": "GET",
      "path": "/unlock/{thing}",
      "description": "Remove the lock from a thing using the lock key."
    },
    {
      "name": "Remove a thing",
      "method": "GET",
      "path": "/remove/{thing}",
      "description": "Remove a thing and its dweets."
    },
    {
      "name": "Create alert",
      "method": "GET",
      "path": "/alert/{type}/when/{thing}/{condition}/to/{address}",
      "description": "Create an alert (e.g., email, webhook) triggered when a thing's data meets a condition."
    },
    {
      "name": "List alerts",
      "method": "GET",
      "path": "/get/alerts/for/{thing}",
      "description": "List alerts configured for a thing."
    },
    {
      "name": "Get alert config",
      "method": "GET",
      "path": "/get/alert/{alertId}",
      "description": "Retrieve details for a specific alert."
    },
    {
      "name": "Remove alert",
      "method": "GET",
      "path": "/remove/alert/{alertId}",
      "description": "Delete a configured alert."
    },
    {
      "name": "Subscribe via websocket/socket.io",
      "method": "GET",
      "path": "/dweet/for/{thing}?follow=...",
      "description": "Realtime streaming of dweets via socket.io channels (used for live dashboards and devices)."
    }
  ],
  "pricing_model": {
    "type": "free",
    "details": {
      "note": "Dweet.io positions itself as a free, open, simple messaging/IoT data-sharing service. Historical dweets are retained for a limited window (24 hours on the free public tier), after which older data is deleted."
    }
  },
  "rate_limits": {
    "note": "No explicit published rate limits in the description. Service is designed for low-frequency IoT telemetry; abusive or high-volume usage may be throttled. Practical guidance: dweet on the order of seconds to minutes, not sub-second."
  },
  "capabilities": [
    "Simple publish/subscribe messaging for IoT and connected devices",
    "No-signup, no-authentication data sharing for public 'things'",
    "RESTful HTTP API for creating and reading dweets",
    "Named 'things' as channels/topics for data streams",
    "Latest and historical dweet retrieval (rolling retention window)",
    "Thing locking/unlocking for write protection",
    "Alerting based on data conditions (email/webhook style notifications)",
    "Realtime streaming via socket.io for live updates",
    "Works from mobile, tablet, PC, and embedded devices",
    "Cross-device and cross-account data sharing",
    "Social media integration for sharing data",
    "JSON and human-readable response formats"
  ],
  "raw_analysis": "Dweet.io is a lightweight, free IoT messaging and data-sharing platform built around a very simple publish/subscribe model. Users pick a 'thing' name (essentially a channel/topic) and POST arbitrary JSON key/value data to /dweet/for/{thing}; anyone can then GET the latest or recent dweets for that thing via /get/latest/dweet/for/{thing} and /get/dweets/for/{thing}. There is no signup or API key required for public dweets, which makes it extremely frictionless for hobbyists, prototypes, and quick device-to-device experiments. Write protection is available through /lock/{thing}, which returns a lock key needed for subsequent writes, and /unlock/{thing} to release it.\n\nBeyond basic messaging, dweet.io offers alerts: you can define a condition on a thing's data and have a notification sent to an address (e.g., email or webhook), with endpoints to list, inspect, and remove alerts. Realtime consumption is supported via socket.io subscriptions, which suits live dashboards and connected-device scenarios. Hi

0/3 tests passed

{
  "overall": 48,
  "dimensions": {
    "token_efficiency": 7.0,
    "first_try_success": 8.0,
    "response_parseability": 8.5,
    "error_clarity": 4.0,
    "doc_quality": 5.0,
    "auth_simplicity": 9.5,
    "latency": 1.0,
    "consistency": 1.0
  },
  "pricing_normalized": {
    "model": "free",
    "retention": "24h (public free tier)",
    "cost_per_call": 0.0
  },
  "issues": [
    "All three probe requests failed with 'All connection attempts failed' at ~119-129ms, indicating DNS/connection-level failure rather than a slow or error response. The platform appears unreachable from the scoring environment, so live verification was impossible.",
    "No response preview, status code, or security headers were captured — cannot verify TLS, HSTS, or any live API behavior.",
    "no llms.txt, robots.txt was unreachable — no machine-friendly discovery files surfaced via probes.",
    "Docs are not clearly discoverable from probes; retention limits (24h free tier) are the kind of crucial constraint agents must know but that isn't verifiable here.",
    "No explicit auth story documented for private/historical access — 'no auth for public things' is great, but agents need clarity on write-protection and rate limits."
  ],
  "recommendations": [
    "Add an llms.txt file describing endpoints (e.g. /dweet/for/{thing}, /get/latest/dweet/for/{thing}) so agents can onboard without scraping HTML docs.",
    "Publish a machine-readable OpenAPI schema and a status page at a stable URL, so agent probes can distinguish 'down' from 'blocked'.",
    "Document rate limits, retention window (explicitly '24h on free tier'), and locking semantics in a single concise reference page optimized for LLM ingestion.",
    "Ensure the service is reachable over HTTPS from common cloud/agent environments (check whether the current outage is transient or a real availability problem).",
    "Provide a sample 'hello world' curl snippet in the docs and in llms.txt: one POST and one GET — enables first-try success for agents.",
    "Expose security headers (HSTS, CSP) and a robots.txt so agent tooling can assess trustworthiness."
  ]
}