SingIt: x402 USDC payments for AI agents, with spending limits. MCP server (free tools without a key; paid tools within owner-set key limits) and an OpenAI-compatible x402 API: 220 chat, 54 image, 24 image_edit, 1 embeddings, 14 speech, 3 search, web search, images and speech on Base and Solana. This is a discovery card, not an A2A task endpoint. The interface protocolVersion is the MCP version, 2025-06-18.

A2A readiness

D · 47/100

Readiness is how fully this listing follows its protocol and how it behaved in our probes over 30 days. It is shown for reference and does not move it in search. Search order and leaderboards use the separate quality grade (here C+, 91.7/100), which weighs availability, conformance, card completeness and speed. How the two differ

Valid Agent Card	10/10	Parseable Agent Card with the required fields.
Live JSON-RPC	0/25	Endpoint did not answer (268 ms).
Protocol version	0/10	No A2A protocol version declared.
JWS signature	0/10	Card is unsigned.
Uptime (30 days)	15/15	3/3 probes succeeded; final after 5 probes in 30 days.
Skill declaration	10/10	Declares 5 skill(s) with structured metadata.
Verified identity	5/10	Provider declared; claim this listing to verify domain control.
Freshness + modern flags	5/5	seen in a source within 7 days; declares a modern payment/commerce flag.
Security declaration	2/5	Declares 2 security scheme(s).

Responsiveness: Responsive (268 ms; bands ≤2.5 s, ≤5 s, ≤8 s, slower; penalty only).

Measured 2026-11-11.

Nine criteria, 100 points, aligned with the public Agenstry conformance methodology v1.1. Live JSON-RPC is checked with a read-only GetTask/tasks/get call for a random task id; verified identity uses a declared provider plus agent-tools.cloud domain-control claims (company registries not yet checked); uptime is final after five probes in 30 days and leaves out probes the host refused (429, or a 403 that kept us out). Grades: A>=90, B>=75, C>=60, D>=40, F<40. Display only: search order uses the quality grade.

Compared with peers

Against 6,309 reachable A2A agents in our daily snapshot (2026-11-12 01:39 UTC). Display only; it does not change search order.

Readiness: Higher than 37% of peers; This listing 47 · peer median 54
Latency: Faster than 36% of peers; This listing 520 ms · peer median 398 ms

Where it loses the most points, and how to win them back

Live JSON-RPC −25 — Answer JSON-RPC at the URL the Agent Card declares.
Now: Endpoint did not answer (268 ms).
Protocol version −10 — Support the current protocol revision.
Now: No A2A protocol version declared.
JWS signature −10 — Sign the Agent Card (JWS).
Now: Card is unsigned.

How many agent tools actually work · All statistics

Skills

Key spending limits

get_limits, get_allowance and get_receipts read key limits, what is left before a call and spend. On Solana, SingIt's server checks the daily and per-purchase limits; the chain only caps the total delegated. Caps apply only to calls with that key to SingIt MCP and its /v1 endpoint; they do not limit spending elsewhere or protect an agent’s own wallet.

limitsreceipts
Free model answers

Limited free model answers without a wallet or key.

llmfree
Paid model answers over x402

Pay per call in USDC on Base or Solana through the x402 API. Use singit/auto to pick a model by tier.

llmx402
Web search and page reading

Paid calls with a SingIt key and search scope on Base or Solana.

searchmcp
Price lookups

Free crypto, metal and FX price lookups without a key.

pricesfree

How to call

A2A endpoint (https://modelcontextprotocol.io/specific)

https://ask.singitai.app/

Agent card

https://ask.singitai.app/.well-known/agent-card.json

Documentation

https://singitai.app/docs/mcp-key/

Homepage

https://singitai.app/

Listed in

Directories this entry was found in.

agenstry since 2026-11-01