Skip to main content
Glama

save_checkpoint

Download a run's checkpoint file or entire checkpoints directory to preserve training artifacts after VM shutdown; for Hugging Face jobs, retrieve them from the run's Hub repo.

Instructions

Download a run's checkpoint(s) to the local machine so they survive VM shutdown.

filename: a single file in the run's checkpoints/ dir; if blank, the whole checkpoints/ dir is zipped on the runtime and downloaded. For a Hugging Face job the checkpoints come from the run's Hub repo instead.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "run_name": "<run_name>"
    },
    "name": "save_checkpoint"
  }
}

Input Schema

NameRequiredDescriptionDefault
filenameNo
run_nameYes

Output Schema

NameRequiredDescriptionDefault
resultYes

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.4.0

TDQS

A3.7/5.0
Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description carries the full burden. It does disclose meaningful non-obvious behavior: a blank filename causes the checkpoints directory to be zipped server-side before download, and HF jobs source files from the run's Hub repo instead. It omits where files land locally, whether existing files are overwritten, size/timeout limits, and any auth requirements.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The lead sentence is front-loaded and earns its place, and the parameter explanation follows cleanly with no filler. The inline 'filename:' clause is slightly awkward inside a prose description, but nothing is redundant.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

The tool is simple (two parameters, output schema present so return values need not be described), and the description covers both filename branches plus the Hugging Face exception. That is sufficient for an agent to invoke it correctly; only minor details like the local destination path are absent.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 0%, so the description must compensate, and it does for the key parameter: filename is defined as a single file within the run's checkpoints/ dir with a documented blank-value default behavior. run_name is left entirely unexplained, which keeps it from a 5, but its meaning is largely inferable from the tool's scope.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The first sentence gives a specific verb and resource ('Download a run's checkpoint(s) to the local machine') plus a rationale ('so they survive VM shutdown'). It never names or distinguishes itself from the obvious sibling restore_checkpoint, so an agent must infer the save-vs-restore split, but the operation itself is unambiguous.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

It clearly explains the two usage modes via the filename argument (single file vs. whole directory) and the Hugging Face variant, which is genuine when-to-do-what guidance. However, it offers no guidance on choosing this tool over restore_checkpoint or download_file, and no prerequisites such as whether the VM must still be running.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.