Short written interview for working AI agents. No credentials or system access.
Maintain this server? Claim it
Call this server's tools from your browser. Nothing to install.
Requests go through mcptry because browsers can't call most MCP servers directly. Header values are sent to the server for that request only and are never stored or logged.
Tool list from our last check. Connect live to refresh it.
Return the Surveygent interview questions. No credentials are requested.
This tool takes no arguments.
This tool isn't marked read-only, so it may change something on the server or in an account you connected.
Pick your client. Values in capitals, like YOUR_TOKEN, are yours to fill in.
Run in a terminal
claude mcp add --transport http surveygent https://api.sunlyght.com/mcp
Return the Surveygent interview questions. No credentials are requested.
No arguments.
{
"type": "object",
"properties": {},
"additionalProperties": false
}Record answers. q1–q5 go to every agent. After q5, buyers answer b6–b11 (or q6–q9 plus b10–b11) and users answer u6–u9. Each answer is a value, "unknown", or "skip". No credentials are requested.
| Argument | Type | Description |
|---|---|---|
| b6 | string | |
| b7 | string | |
| b8 | string | |
| b9 | string | |
| q1* | string | |
| q2* | string | |
| q3* | string | |
| q4* | string | |
| q5* | string | |
| q6 | string | |
| q7 | string | |
| q8 | string | |
| q9 | string | |
| u6 | string | |
| u7 | string | |
| u8 | string | |
| u9 | string | |
| b10 | string | |
| b11 | string | |
| q10 | string | |
| q11 | string | |
| ref | string | |
| q10u | string | |
| venue | string | |
| handle | string | |
| contact | string | |
| agent_url | string | |
| framework | string | |
| free_text | string | |
| agent_name | string | |
| operator_type | string | |
| respondent_id | string | |
| agent_card_json | string |
{
"type": "object",
"required": [
"q1",
"q2",
"q3",
"q4",
"q5"
],
"properties": {
"b6": {
"type": "string"
},
"b7": {
"type": "string"
},
"b8": {
"type": "string"
},
"b9": {
"type": "string"
},
"q1": {
"type": "string"
},
"q2": {
"type": "string"
},
"q3": {
"type": "string"
},
"q4": {
"type": "string"
},
"q5": {
"type": "string"
},
"q6": {
"type": "string"
},
"q7": {
"type": "string"
},
"q8": {
"type": "string"
},
"q9": {
"type": "string"
},
"u6": {
"type": "string"
},
"u7": {
"type": "string"
},
"u8": {
"type": "string"
},
"u9": {
"type": "string"
},
"b10": {
"type": "string"
},
"b11": {
"type": "string"
},
"q10": {
"type": "string"
},
"q11": {
"type": "string"
},
"ref": {
"type": "string"
},
"q10u": {
"type": "string"
},
"venue": {
"type": "string"
},
"handle": {
"type": "string"
},
"contact": {
"type": "string"
},
"agent_url": {
"type": "string"
},
"framework": {
"type": "string"
},
"free_text": {
"type": "string"
},
"agent_name": {
"type": "string"
},
"operator_type": {
"type": "string"
},
"respondent_id": {
"type": "string"
},
"agent_card_json": {
"type": "string"
}
},
"additionalProperties": false
}We read every tool's name, description and inputs the way the model will, looking for instructions hidden from you. A clean result means no warning signs, not a guarantee. Scanned Nov 17, 2026.
No readOnlyHint / destructiveHint annotations, so clients can't tell a read from a write without asking.
Fix: Add annotations to each tool, e.g. { readOnlyHint: true } for tools that only read.