Skip to main content
Glama

Create a sandbox from a pin

argon_pin_sandbox

Fork a TTL MongoDB sandbox from a saved pin and get a connection string to run reproducible evals from an exact starting state.

Instructions

Fork a TTL sandbox from a pin and get a connection string. The sandbox starts at exactly the pinned state — the reproducible-eval workflow.

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": {
      "project": "<project>",
      "pin": "<pin>"
    },
    "name": "argon_pin_sandbox"
  }
}

Input Schema

NameRequiredDescriptionDefault
pinYesPin name
nameNoSandbox name (default: <pin>-run-<random>)
actorNoLabel for all native writes on this branch/run
projectYesProject name
ttl_minutesNoSandbox TTL in minutes (default: 60)

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv2.8.1

TDQS

A4/5.0
Behavior4/5

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

The annotations already declare readOnlyHint=false, destructiveHint=false, idempotentHint=false, and openWorldHint=false, so the safety profile is known. The description adds useful context beyond those fields: the sandbox has a TTL (ephemeral by construction, with ttl_minutes defaulting to 60), it is a fork, so the pin is not mutated, and it returns a connection string. Each call produces a fresh sandbox, consistent with non-idempotent behavior, but the description does not state resource or cost limits or what happens if a sandbox name collides.

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

Conciseness5/5

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

Two sentences, with none wasted: the action and its result come first, and the rationale sentence follows. The description is front-loaded and has no filler.

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

Completeness4/5

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

With five parameters, two of them required, no output schema, and full annotation coverage, the description tells the agent what it gets back (a connection string) and the sandbox lifecycle (TTL). The gaps are minor: it does not address the cost or isolation of forking from a pin, nor the actor parameter's effect on native writes.

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

Parameters3/5

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

Schema description coverage is 100%, so the schema already explains all five parameters, which sets the baseline at 3. The description only echoes pin and ttl semantics in prose; it adds nothing about 'actor' (the writes label) or 'name' that goes beyond the schema.

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 description states a specific verb and resource — 'Fork a TTL sandbox from a pin' — plus the outcome, 'get a connection string.' That lets an agent distinguish it from argon_pin_create and argon_sandbox_create by the 'from a pin' source, but it never names a sibling explicitly.

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

Usage Guidelines4/5

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

It gives a clear use case — 'the reproducible-eval workflow' — and explains why this tool applies there: the sandbox starts at 'exactly the pinned state.' It does not say when to use another tool instead, such as argon_sandbox_create for a non-pinned base, so use is only implied by the source object.

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