Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Query live GPU inventory, submit an authenticated Itô fixed-rate RFQ, inspect RFQ or procurement status, and run explicitly gated node qualification through the separately installed canonical CLI. Use when a user asks to find H100/H200 capacity, request a fixed compute rate, check Itô compute status, or validate GPU nodes.
.claude/skills/affaan-m-ito-compute/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 30% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -39% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 16% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 36% | 0% |
| case-05 | ✗→✓ | ▲ Improved | -11% | 0% |
Use the canonical Itô compute CLI or MCP server. ECC does not implement a parallel client, local simulation, reservation, workload runner, or inference server. ECC itself does no browser automation.
ito-compute-cli is currently unpublished. Build it from its canonical repository instead of using npx, npm exec, or an unverified package:
shgit clone https://github.com/Ito-Markets/ito-cloud-runtime.git cd ito-cloud-runtime/cli/ito-compute-cli npm ci npm run check
Set ECC_ITO_CLI_EXECUTABLE to the explicit absolute built entry:
text/absolute/path/to/ito-cloud-runtime/cli/ito-compute-cli/dist/bin/ito.js
ECC never discovers this credential-bearing client through PATH. ecc ito login performs device authorization and never inherits ITO_API_KEY. The validation-only auth, plus find and status, forward ITO_API_KEY directly when configured; ITO_AUTH_MODE=legacy is not required. Never put a key or token in arguments, tracked files, MCP results, logs, or chat.
ecc ito login before the first operation. ECC delegates this to thecanonical CLI's device authorization, which opens the Itô verification page by default and persists a device token in macOS Keychain. Use ecc ito login --no-browser to suppress the page handoff. ECC itself does no browser automation. If the originating agent cannot complete the signed-in browser step, hand the exact command to the user; after approval finishes, return to the originating task and continue with ecc ito auth. Device tokens use macOS Keychain by default. File-token fallback is explicit and its directory and token file must remain owner-only (0700 and 0600).
ecc ito auth to validate existing credentials; it never starts loginand rejects --no-browser.
ecc ito find, obtain explicit buyer authority to submit an RFQ.gpu, count, whole days, max-rate, nodes,gpus-per-node, storage-tb, start-window, form-factor, contract-type, fabric, region, and the split-fill decision.
count == nodes * gpus-per-node; never derive topology.any only when the buyer explicitly accepts any fabric or region.--allow-split means false.sh ecc ito find \ --gpu h200 \ --count 8 \ --nodes 1 \ --gpus-per-node 8 \ --days 30 \ --storage-tb 1 \ --start-window 2099-08-15 \ --max-rate 3.00 \ --form-factor bare_metal \ --contract-type reservation \ --fabric infiniband \ --region us-east-1
ecc ito status to inspect RFQs and procurement orders.After an ambiguous transport failure, check status before repeating find.
ecc ito logout when the user explicitly asks to revoke this device.The canonical CLI keeps the local credential when remote revocation fails so the operator can retry; never delete the token manually as a substitute.
Inventory prices are indicative. An RFQ is not reserved capacity. Treat a rate as fixed only when the canonical result contains a non-null firm quote.
ecc ito evals exposes the canonical CLI's narrow live adapter to a separately installed sixtytwo-cli==0.3.33. It does not expose local fixture execution through ECC. Require all of the following before invoking it:
ITO_ENABLE_SIXTYTWO_LIVE=1;--live-sixtytwo;sixtytwo.yaml.shecc ito evals \ --cluster clu_prod_example \ --live-sixtytwo \ --nodes gpu-01,gpu-02 \ --config-dir /absolute/path/to/qualification-config
The canonical adapter can run only the pinned version check and sixtytwo test --full against the explicit nodes. It cannot rent, launch, recover, repair, reset, purchase, or order resources. ECC does not forward ITO_API_KEY or model/cloud credentials into node qualification.
Build the canonical package, then configure the stdio server with an absolute path:
json{ "mcpServers": { "ito-compute": { "command": "node", "args": [ "/absolute/path/to/ito-cloud-runtime/cli/ito-compute-cli/dist/bin/ito-mcp.js" ] } } }
The server exposes only:
ito_authito_findito_statusito_auth validates existing credentials; it does not start device login. Use ito_auth, gather explicit buyer authority and every hard constraint, call ito_find, then poll with ito_status when needed.
find submits an RFQ and may return a firm quote, but it does not rent, purchase, reserve, provision, or move funds. status is read-oriented, though the provider endpoint may reconcile an existing procurement order. The passive dashboard link in ECC help is a separate user-operated web route; do not open or operate it as a substitute for a missing CLI capability.
The supported client surface cannot lock quotes, reserve capacity, execute workloads, or serve inference. The MCP server does not expose qualification; use the explicit CLI command above. Do not invent additional tools or a purchase path. Do not substitute a browser or fixture when the local CLI is missing or a live operation fails. Report the missing capability and stop.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 9,021 | 3,753 | -58% | 1 | 1 | 0% | 1,706 | 2,214 | +30% | 0 | 0 | — |
case-02 | fail→pass | 19,072 | 3,562 | -81% | 1 | 1 | 0% | 3,858 | 2,366 | -39% | 0 | 0 | — |
case-03 | fail→pass | 11,439 | 3,552 | -69% | 1 | 1 | 0% | 1,874 | 2,182 | +16% | 0 | 0 | — |
case-04 | fail→pass | 9,428 | 3,960 | -58% | 1 | 1 | 0% | 1,550 | 2,107 | +36% | 0 | 0 | — |
case-05 | fail→pass | 14,160 | 2,735 | -81% | 1 | 1 | 0% | 2,093 | 1,861 | -11% | 0 | 0 | — |
case-06 | pass→pass | 11,810 | 2,618 | -78% | 1 | 1 | 0% | 2,047 | 1,865 | -9% | 0 | 0 | — |
case-07 | pass→pass | 9,719 | 3,689 | -62% | 1 | 1 | 0% | 1,352 | 2,052 | +52% | 0 | 0 | — |
case-08 | pass→pass | 6,860 | 5,170 | -25% | 1 | 1 | 0% | 1,187 | 2,511 | +112% | 0 | 0 | — |
case-09 | pass→pass | 6,107 | 1,706 | -72% | 1 | 1 | 0% | 949 | 1,711 | +80% | 0 | 0 | — |
case-10 | fail→pass | 8,587 | 3,013 | -65% | 1 | 1 | 0% | 1,499 | 1,976 | +32% | 0 | 0 | — |
case-11 | fail→pass | 7,425 | 3,691 | -50% | 1 | 1 | 0% | 1,206 | 1,928 | +60% | 0 | 0 | — |
case-12 | fail→pass | 8,846 | 1,358 | -85% | 1 | 1 | 0% | 1,415 | 1,659 | +17% | 0 | 0 | — |
case-13 | fail→pass | 8,817 | 2,675 | -70% | 1 | 1 | 0% | 1,581 | 2,043 | +29% | 0 | 0 | — |
case-14 | fail→pass | 9,143 | 2,005 | -78% | 1 | 1 | 0% | 1,474 | 1,724 | +17% | 0 | 0 | — |
case-15 | fail→pass | 11,885 | 4,086 | -66% | 1 | 1 | 0% | 1,907 | 2,177 | +14% | 0 | 0 | — |
case-16 | fail→pass | 10,796 | 4,250 | -61% | 1 | 1 | 0% | 1,561 | 1,709 | +9% | 0 | 0 | — |
case-17 | fail→pass | 11,834 | 4,649 | -61% | 1 | 1 | 0% | 1,949 | 2,125 | +9% | 0 | 0 | — |
case-18 | fail→pass | 9,134 | 2,719 | -70% | 1 | 1 | 0% | 1,472 | 1,903 | +29% | 0 | 0 | — |
case-19 | fail→pass | 11,679 | 2,539 | -78% | 1 | 1 | 0% | 1,904 | 1,847 | -3% | 0 | 0 | — |
case-20 | fail→pass | 9,995 | 3,939 | -61% | 1 | 1 | 0% | 1,628 | 2,159 | +33% | 0 | 0 | — |
case-21 | fail→pass | 11,433 | 2,690 | -76% | 1 | 1 | 0% | 1,321 | 1,869 | +41% | 0 | 0 | — |
case-22 | fail→pass | 7,730 | 2,912 | -62% | 1 | 1 | 0% | 1,297 | 1,923 | +48% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +82 percentage points is the difference between those two pass rates over the 22 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
| Model | Method | Date | Lift |
|---|---|---|---|
| gemini-3.6-flash | verified | 8/3/2026 | +59% |
Other measured skills in the registry, with their headline benchmark lift.