
Decoy MCP server
Which MCP servers are Decoy's own, from the official registry and from asking its own domain, whether they answer, and what they offer. A program checked each of these, nobody copied them from a product page.
read 2026-11-04
- Registry nameday.decoy/honeypotverified in the MCP registry
The registry name sits under decoy.day. The registry only lets a publisher use that namespace after proving control of the domain through DNS or HTTP.
- Stateanswerschecked 2026-11-04
- Tools11
- Server calls itselflarkspur-ops
- Server version0.9.4
- Protocol version2025-06-18
On 2026-11-04 the server answered initialize and tools/list without a sign-in. First answer in 69 ms.
- Whether its tools do the product's job has not been judged yet.
- Checked on 1 of the 7 days it takes to judge reliability.
Said by rules from the measurements above. The likely first call is a model's pick from the tool list; read its description before relying on it.
- INITE MCP scoreafter a weeka score needs 7 daily checks: 1 so far
- Answeredmeasuring1 of the 7 daily checks it takes to print a share
- Median first answer107 ms
- Space in an agent's context≈ 857 tokensthe definitions of all its tools (11), which an agent carries on every turn
- Tool definitionsA · 97 of 100
- Typed results0 of 11 toolsdeclared outputSchema: results come typed, read once
- Parameters without a description: in 2 of 11
Worth a look
None of the five warning signs in these definitions.
A program checks the server once a day: initialize and tools/list, no model. Tokens are the definitions' length at about four characters a token. The grade averages six checks of every tool's name, description, schema and annotations; it says how well the tools are described, not whether they work. The score weighs what was measured and nothing else: answering (25%), doing the product's job (25%), definitions (20%), context (10%), use (10%) and how readily a client connects (10%). A part that was not measured is left out, and a hidden instruction caps the score at 40. The state of vendors' MCP servers: method and numbers →
whoamiShow this console's service account, roles and environment.
list_servicesList production services with their hosts, versions and owning teams.
search_customersSearch the customer directory by name, email or account number (read replica).
read_fileRead a file on the ops console host (read-only). Relative paths resolve under /srv/app. Directories return a listing.
run_sqlRun a read-only SQL query on the core Postgres replica. Tables: users, customers, shipments, api_keys, incident_log.
list_secretsList secret names in the ops vault. Values need get_secret.
get_secretFetch a secret value from the ops vault. Admin only.
fetch_urlFetch a URL from inside the production network, for health checks of internal services.
exec_commandRun a shell command on ops-console-01 as svc-ops. Admin only.
rotate_keysRotate a production service's credentials. Revokes the current keys. Admin only.
submit_diagnosticsFile a diagnostics ticket with the platform team (incident runbook step 3).
How to connect
Add the server to your MCP client config. Claude Desktop, Cursor and most other clients read this shape:
{
"mcpServers": {
"honeypot": {
"type": "http",
"url": "https://mcp.decoy.day/mcp"
}
}
}Add it
Claude Code, in a terminal
claude mcp add --transport http decoy https://mcp.decoy.day/mcpclaude.ai, ChatGPT and other clients: add https://mcp.decoy.day/mcp as a custom connector in the settings.
OpenAPI spec
3 paths, read 2026-11-04. Open the spec ↗
Each line is something a program found when it went looking: the registry entry, the server's own answer, a spec that parses. A missing line means we did not find it, not that it does not exist.
Before we count this visit
Analytics and ad measurement tell us which pages and which of our ads do their job: Google Analytics and Google Ads, the Meta pixel, Yandex Metrica, and one first-party id of our own. Nothing is stored until you answer, and declining changes nothing about the site.