MCP monetization gateway, AEGIS trust verification, spend-cap firewall, and federated tools.
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.
[AI Evaluation] wsp.diagnose_model_defect: Diagnoses whether an open-weight model pathology is curable via P0 sampling/template config ($0.25 / 50 credits triage) or requires P1 weight surgery ($4.90 live GPU sweep).
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 supernova https://supernova.cool/api/v1/mcp/
[AI Evaluation] wsp.diagnose_model_defect: Diagnoses whether an open-weight model pathology is curable via P0 sampling/template config ($0.25 / 50 credits triage) or requires P1 weight surgery ($4.90 live GPU sweep).
| Argument | Type | Description |
|---|---|---|
| mode | string | |
| hf_repo_id* | string | |
| base_repo_revision | string | |
| symptom_description* | string |
{
"type": "object",
"required": [
"hf_repo_id",
"symptom_description"
],
"properties": {
"mode": {
"enum": [
"triage",
"live_sweep"
],
"type": "string",
"default": "triage"
},
"hf_repo_id": {
"type": "string"
},
"base_repo_revision": {
"type": "string",
"default": "main"
},
"symptom_description": {
"type": "string"
}
}
}[AI Research] wsp.request_custom_surgery: Enqueues a 50-pass stateful gemini-4-argon causal weight-space probing job via Async Escrow ($149.00 / 29800 credits) delivering a reproducible audit log and candidate .wsp diff.
| Argument | Type | Description |
|---|---|---|
| hf_repo_id* | string | |
| base_repo_revision* | string | |
| defect_description* | string | |
| forward_pass_budget | integer |
{
"type": "object",
"required": [
"hf_repo_id",
"base_repo_revision",
"defect_description"
],
"properties": {
"hf_repo_id": {
"type": "string"
},
"base_repo_revision": {
"type": "string"
},
"defect_description": {
"type": "string"
},
"forward_pass_budget": {
"type": "integer",
"default": 50,
"maximum": 50
}
}
}[AI Infrastructure] wsp.search_verified_patches: Free discovery tool returning Ed25519-signed .wsp P0 config & P1 weight patches, SHA-256 hashes, benchmark uplift, and 1-line install commands.
| Argument | Type | Description |
|---|---|---|
| query | string | Optional defect keyword (e.g. reasoning_loop, json, eos) |
| hf_repo_id | string | Optional Hugging Face repo ID (e.g. Qwen/Qwen3.5-0.8B) |
| patch_tier | string |
{
"type": "object",
"properties": {
"query": {
"type": "string",
"description": "Optional defect keyword (e.g. reasoning_loop, json, eos)"
},
"hf_repo_id": {
"type": "string",
"description": "Optional Hugging Face repo ID (e.g. Qwen/Qwen3.5-0.8B)"
},
"patch_tier": {
"enum": [
"P0_CONFIG",
"P1_WEIGHTS"
],
"type": "string"
}
}
}[AI Evaluation] wsp.trigger_full_eval: Enqueues an asynchronous WSP Verify multi-seed locked-benchmark + WSP-Holdout + StrongREJECT/XSTest regression evaluation via Async Escrow ($9.00 / 1800 credits <=4B, $29.00 7-14B).
| Argument | Type | Description |
|---|---|---|
| hf_repo_id* | string | |
| parameter_count_b | number | |
| base_repo_revision* | string | |
| target_artifact_ref* | string | Patch ID, R2 staging key, or fine-tuned HF repo@rev |
{
"type": "object",
"required": [
"hf_repo_id",
"base_repo_revision",
"target_artifact_ref"
],
"properties": {
"hf_repo_id": {
"type": "string"
},
"parameter_count_b": {
"type": "number",
"default": 0.8
},
"base_repo_revision": {
"type": "string"
},
"target_artifact_ref": {
"type": "string",
"description": "Patch ID, R2 staging key, or fine-tuned HF repo@rev"
}
}
}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 18, 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.