AiAuN
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| resolve_experiment_requestB | Check required experiment fields. If missing is non-empty, ask the user; do not run other tools yet. |
| inspect_local_dirA | Inspect a local data directory: file count, total bytes, sample names. |
| list_repo_configsA | List experiment YAML configs under semi-mask2former/configs. |
| kaggle_dataset_checkA | Check whether a Kaggle dataset owner/slug already exists. |
| kaggle_dataset_pushC | Upload a local folder as a Kaggle dataset. Size is limited only by the Kaggle API. |
| kaggle_kernel_pushC | Push a Kaggle script kernel. dataset_slugs is a comma-separated owner/slug list. |
| kaggle_kernel_statusC | Poll Kaggle kernel execution status. |
| kaggle_kernel_logsB | Fetch kernel output logs (often only after complete). |
| aiaun_smoke_scriptA | Return the inlined synthetic color-class smoke kernel source (no secrets). |
| aiaun_resnet50_gpu_smoke_scriptC | ResNet50 GPU smoke: short train, inference on best.pt, Drive upload of artifacts (no secrets in source). Push with enable_gpu=true. |
| experiment_tracking_linksC | Return Kaggle / W&B / Drive URLs the user can open to track this experiment. Always show these links in the chat. |
| drive_folder_infoB | Verify Google Drive service-account access to GOOGLE_DRIVE_FOLDER_ID. |
| drive_upload_fileB | Upload a file to the configured Drive folder (Shared Drive + SA; no browser). |
| kaggle_kernel_output_to_driveA | Download Kaggle kernel Output then upload to Drive via the local service account (SSH-safe, no browser). Prefer kernel-side User Secrets upload when possible. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
| run_kaggle_experiment | Workflow for running an experiment on Kaggle via this MCP. |
| generate_experiment_notebook | Guide for writing a Kaggle script that clones GitHub and uses User Secrets. |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 14 tools
Each tool targets a distinct step in the Kaggle/Drive experiment workflow: validation, local inspection, dataset/kernel operations, script source, tracking, and Drive transfer. Only the two smoke-script tools are close variants, but their names and descriptions clearly separate color-class vs. ResNet50 GPU usage.
Names are readable but grammatically inconsistent: some are verb-first like inspect_local_dir, some are object-verb like kaggle_kernel_push, and others are noun phrases like experiment_tracking_links or drive_folder_info. The kaggle_* and drive_* prefixes provide some grouping, but there is no single predictable verb_noun convention.
14 tools is a reasonable size for an experiment pipeline spanning local inspection, Kaggle dataset/kernel lifecycle, and Drive output transfer. Each tool maps to a needed step, so the count feels well-scoped.
The set covers the main loop: validate request, inspect data, push dataset, push kernel, poll logs, move outputs to Drive, and provide tracking links. Minor gaps exist—configs can be listed but not read, and there is no update/cancel for kernels—but agents can generally work around them.