cloudgpu-mcp 0.16.4 → 0.16.6
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.md +3 -1
- package/package.json +2 -2
- package/skills/cloudgpu/SKILL.md +1 -1
- package/src/index.js +6 -4
- package/src/regional-offers.js +1 -1
package/README.md
CHANGED
|
@@ -179,6 +179,8 @@ The server also provides a `train_on_cloudgpu` prompt.
|
|
|
179
179
|
|
|
180
180
|
## Guardrails
|
|
181
181
|
|
|
182
|
+
- **Bounded local MCP inputs.** A tool call may contain at most 10,000 combined array entries and object members, including nested values. Larger calls are rejected before tool-schema validation or API/SSH handlers. This includes inline ComfyUI workflows; simplify oversized workflows rather than splitting a paid operation or silently dropping fields. The stdio transport also retains its 10 MiB buffered-input limit. These are local MCP limits, not a claim about the remote service.
|
|
183
|
+
|
|
182
184
|
- **A quote first.** `create_instance` only takes a `quote_id` from `estimate_cost` and `user_confirmed=true`.
|
|
183
185
|
- **Server-side limits.** Every instance is created with `max_spend_usd`, `max_hours` and an idle limit (60 minutes by default). CloudGPU's servers check them and destroy the instance when one is hit, even if your client has disconnected or your laptop is asleep.
|
|
184
186
|
- **Your own ceiling.** `CLOUDGPU_MAX_SPEND_USD` and `CLOUDGPU_MAX_HOURS` cap what the agent may set.
|
|
@@ -469,7 +471,7 @@ Check anonymous `GET /v1/public/rental-regions` (MCP `list_rental_regions`) firs
|
|
|
469
471
|
|
|
470
472
|
For a rental, pass the same `continent` to `estimate_cost` / `POST /v1/gpu/billing-quote`, then obtain explicit approval and pass it with the exact `billing_quote_id` and `billing_accepted: true` to `POST /v1/gpu/instances`. The stdio client's `create_instance` carries the approved continent automatically. Changes or missing continent bindings require a new quote. Minimum billing periods are unchanged; there is no automatic cross-continent fallback.
|
|
471
473
|
|
|
472
|
-
The generic `POST /v1/quotes`, Jobs and batches do not yet accept a hard continent constraint. Do not drop that constraint or substitute an unconstrained Job.
|
|
474
|
+
The generic `POST /v1/quotes`, Jobs and batches do not yet accept a hard continent constraint. Do not drop that constraint or substitute an unconstrained Job. Rental catalogue prices follow qualified source costs and current stock. The lowest available qualifying price is shown, while the accepted billing quote binds the price and billing terms. Native regional observations remain planning data, not reservations or proof of worldwide rentable coverage.
|
|
473
475
|
### Detailed GPU configurations
|
|
474
476
|
|
|
475
477
|
`list_gpus` preserves the REST `/v1/gpu/offers` configuration data. `usd_per_gpu_hour` is a starting per-card rate; each `configurations[].hourlyPrice` is the **whole instance** price for `gpuCount` cards. Each entry includes per-card VRAM, CPU/RAM, system/local disk, interconnect, continent, current availability, billing minimum/increment in seconds and observation time in Unix milliseconds where supplied. Missing fields mean unknown.
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "cloudgpu-mcp",
|
|
3
|
-
"version": "0.16.
|
|
3
|
+
"version": "0.16.6",
|
|
4
4
|
"mcpName": "io.github.cloudgpu-app/cloudgpu-mcp",
|
|
5
5
|
"description": "MCP server and command-line tool for cloudgpu.app: let Claude Code, Cursor, Cline, opencode, goose and other MCP clients (or you, from a terminal) rent a GPU with explicit billing quotes, run a job, fetch the results and shut it down, with spending limits the platform enforces.",
|
|
6
6
|
"type": "module",
|
|
@@ -53,6 +53,6 @@
|
|
|
53
53
|
"dependencies": {
|
|
54
54
|
"@modelcontextprotocol/sdk": "^1.32.2",
|
|
55
55
|
"ssh2": "^1.17.0",
|
|
56
|
-
"zod": "^
|
|
56
|
+
"zod": "^4.6.7"
|
|
57
57
|
}
|
|
58
58
|
}
|
package/skills/cloudgpu/SKILL.md
CHANGED
|
@@ -100,4 +100,4 @@ If installed, `prepare_task_pipeline` prepares up to eight reviewed sequential c
|
|
|
100
100
|
|
|
101
101
|
## Regional discovery
|
|
102
102
|
|
|
103
|
-
Use `list_regional_offers` for recent native regional observations and independent price evidence. Ask for the human owner's country only when useful; do not infer it from the Agent server IP. Respect unknown/expired facts and shared pools. These are planning observations, not reservations, region-bound quotes or verified multi-card allocations.
|
|
103
|
+
Use `list_regional_offers` for recent native regional observations and independent price evidence. Ask for the human owner's country only when useful; do not infer it from the Agent server IP. Respect unknown/expired facts and shared pools. These are planning observations, not reservations, region-bound quotes or verified multi-card allocations. For continent-constrained rentals, check list_rental_regions for enabled=true, then use estimate_cost and the approved billing quote in create_instance. Rental catalogue prices follow qualified supply. Exact countries/datacenters and continent-constrained Jobs remain unsupported; never replace a hard requirement with an unconstrained allocation. Existing R2 Workspace is explicit file save/restore, not a mounted volume.
|
package/src/index.js
CHANGED
|
@@ -42,7 +42,7 @@ import {
|
|
|
42
42
|
shq,
|
|
43
43
|
} from "./guards.js";
|
|
44
44
|
|
|
45
|
-
const VERSION = "0.16.
|
|
45
|
+
const VERSION = "0.16.6";
|
|
46
46
|
const HERE = path.dirname(fileURLToPath(import.meta.url));
|
|
47
47
|
const RULES = fs.readFileSync(path.join(HERE, "..", "rules", "cloudgpu.md"), "utf8");
|
|
48
48
|
|
|
@@ -68,7 +68,7 @@ const quotes = new Quotes();
|
|
|
68
68
|
const session = new Map();
|
|
69
69
|
|
|
70
70
|
const INSTRUCTIONS = `CloudGPU (cloudgpu.app) provides GPU compute scheduling for AI agents from a prepaid balance. Billing is source-specific: per-minute or minimum one hour with each started hour charged in full. Never infer the period from an hourly price label.
|
|
71
|
-
Regional discovery: list_regional_offers reads native observations with separate stock/price ages and explicit unknowns. Supply the owner country explicitly, never the Agent egress IP. Native observations are planning only. For continent-constrained rentals, first call list_rental_regions and require enabled=true, then estimate_cost with the chosen continent. create_instance carries that approved constraint. Countries/datacenters and continent-constrained Jobs are not supported; never drop a hard constraint.
|
|
71
|
+
Regional discovery: list_regional_offers reads native observations with separate stock/price ages and explicit unknowns. Supply the owner country explicitly, never the Agent egress IP. Native observations are planning only. For continent-constrained rentals, first call list_rental_regions and require enabled=true, then estimate_cost with the chosen continent. create_instance carries that approved constraint. Countries/datacenters and continent-constrained Jobs are not supported; never drop a hard constraint. Rental catalogue prices follow currently qualified supply; obtain a fresh billing quote before creating an instance.
|
|
72
72
|
Planning: compare_compute_costs compares billed GPU cost for explicit candidates and duration; it is not a benchmark or creation approval. Check existing list_instances before renting: multiple approved commands can use the same running rental within its existing budget, time and idle limits. Separate run_job calls each provision and release independently; prepare_job_rerun does not reuse a live GPU. Never extend limits or keep idle GPUs merely to consume prepaid time.
|
|
73
73
|
Server batches: prepare_job_batch also prepares the exact manifest for run_server_job_batch. The new tool atomically enqueues on the server, persists across client/server restarts, and requires an aggregate cap and individually approved billing. Use get_server_job_batch/cancel_server_job_batch; never substitute client partial submission when unavailable. No automatic failed-command retries.
|
|
74
74
|
Legacy client batches: prepare_job_batch reviews up to eight individually quoted Jobs and their combined GPU budget. run_job_batch requires an exact approved plan hash; partial submission is possible. Retain the manifest and stable batch key to reconcile lost responses. No automatic execution retries; get_job_batch never proves independent supplier release. Storage charges are additional to GPU caps.
|
|
@@ -80,7 +80,9 @@ Jobs: for a batch task (one command that writes its results somewhere), run_job
|
|
|
80
80
|
No API key yet? connect_account gives the user a link to approve this assistant with a spending limit; then finish_connect stores the key on this computer.
|
|
81
81
|
The same API key also works for the OpenAI-compatible model API at ${cfg.baseUrl}/v1 (see list_models).`;
|
|
82
82
|
|
|
83
|
-
|
|
83
|
+
// Bound collection work before any tool schema validation or handler execution.
|
|
84
|
+
// Includes nested array entries and object members (not string bytes).
|
|
85
|
+
const server = new McpServer({ name: "cloudgpu", version: VERSION }, { instructions: INSTRUCTIONS, maxToolInputElements: 10000 });
|
|
84
86
|
useStrictToolInputs(server);
|
|
85
87
|
registerFundingTools(server,{api,wrap,ok,readOnly:cfg.readOnly});
|
|
86
88
|
registerAgentReadinessTools(server,{api,wrap,ok});
|
|
@@ -531,7 +533,7 @@ if (!cfg.readOnly) {
|
|
|
531
533
|
},wrap(async (input)=>ok(await resolveModelReference(input))));
|
|
532
534
|
server.registerTool("prepare_model_deployment", {
|
|
533
535
|
title:"Prepare models and a ComfyUI workflow",description:"Preparation only; no GPU is started. Create a reviewed setup script for the existing ComfyUI template. Downloads go from the public model host directly to the rented GPU, never through CloudGPU storage. Use model names as categories (checkpoints, loras, vae, diffusion_models, text_encoders, clip, controlnet, upscale_models). Does not install custom nodes, execute workflows, accept licenses for the user, or promise compatibility. Pass the setup_script only after explicit approval to the existing deployment flow.",
|
|
534
|
-
inputSchema:{model_references:z.array(z.record(z.unknown())).min(1).max(4),workflow:z.record(z.unknown()).optional()},
|
|
536
|
+
inputSchema:{model_references:z.array(z.record(z.string(),z.unknown())).min(1).max(4),workflow:z.record(z.string(),z.unknown()).optional()},
|
|
535
537
|
annotations:{readOnlyHint:true,destructiveHint:false,idempotentHint:true,openWorldHint:false},
|
|
536
538
|
},wrap(async (input)=>ok(await api.prepareModelDeployment(input))));
|
|
537
539
|
server.registerTool("prepare_job_rerun", {
|
package/src/regional-offers.js
CHANGED
|
@@ -12,7 +12,7 @@ export function registerRegionalOfferTools(server,{api,wrap,ok}){
|
|
|
12
12
|
|
|
13
13
|
server.registerTool("list_regional_offers",{
|
|
14
14
|
title:"Inspect regional compute observations",
|
|
15
|
-
description:"Anonymous, read-only native regional stock observations, independent price evidence ages, billing references and known/unknown limits. Supply the human owner's country explicitly; the Agent server IP is not its owner's location. Counts are not reservations or verified single-host multi-GPU allocations. Published SKU references are not region-bound executable quotes;
|
|
15
|
+
description:"Anonymous, read-only native regional stock observations, independent price evidence ages, billing references and known/unknown limits. Supply the human owner's country explicitly; the Agent server IP is not its owner's location. Counts are not reservations or verified single-host multi-GPU allocations. Published SKU references are not region-bound executable quotes; these observations are not executable quotes. Use list_rental_regions and estimate_cost for enabled continent-constrained rentals; exact countries/datacenters are unsupported. Unsupported, stale or failed observations do not prove global stockout. Follow has_more/next_after_id; pagination is a live view, not a frozen snapshot.",
|
|
16
16
|
inputSchema:{user_country:z.string().regex(/^[A-Za-z]{2}$/).optional(),country_code:z.string().regex(/^[A-Za-z]{2}$/).optional(),gpu_model:z.string().max(100).optional(),gpu_count:z.number().int().min(1).max(8).default(1),include_unavailable:z.boolean().default(false),limit:z.number().int().min(1).max(200).default(100),after_id:z.string().regex(/^offer_[a-f0-9]{32}$/).optional()},
|
|
17
17
|
annotations:{readOnlyHint:true,destructiveHint:false,idempotentHint:true,openWorldHint:true},
|
|
18
18
|
},wrap(async input=>{
|