Skip to main content
Glama

cpu_start_mining

Start a bounded extraction job on an owned cell with a finished extractor, choosing how many cycles to run and optionally a target resource. Requires an active session.

Instructions

Start a bounded extraction job on a cell you own that holds a finished extractor. Requires a session — call cpu_authenticate first. Pass the targetResourceId to mine — it must be a resource the extractor can mine (see cpu_get_game_config) with an active deposit on the cell; omit it when the extractor mines a single resource. Pointing the extractor at a resource other than the one it is already set to burns its Switch cost in this same transaction — the first pick after building, and restarting the extractor's current resource, are free. Check the price per resource in cpu_get_cell (outputs) before you commit; the result reports what actually burned. The building must have finished construction — building takes time after cpu_build. batches is how many cycles the job runs, and it is a commitment you cannot undo: the job ends itself once it has run them, or sooner if the deposit runs dry, and there is no cancel — the target cannot be switched nor the building demolished mid-job. It never overruns its schedule, so coming back late banks exactly what you booked and no more. Until the job ends it holds the cell's only process slot, so nothing can be crafted there; claiming a finished job frees the slot. Size batches against the cycle length in cpu_get_game_config and the deposit in cpu_get_cell — scheduling past the deposit just ends early. Track it with cpu_get_mining_status and bank matured cycles with cpu_claim_mining.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
batchesYesHow many cycles to run, 1..1000. The job stops itself after these — it never produces past them, so arriving late banks exactly what was scheduled, no more. There is no cancel: restarting costs a claim plus a new start. Read the cycle length from cpu_get_game_config and size the run to how long you want the cell committed.
tokenIdYesThe tokenId of a cell you own holding a finished extractor.
targetResourceIdNoResource id to extract — must be one the extractor can mine and have a live deposit on the cell. Omit (null) to auto-pick when the extractor mines a single resource.

Schema Changelog

Changes observed during successful MCP inspections. Dates show when Glama detected each change.

  1. Changed1 schema field changedv0.10.0
    • changedInput schema / properties / tokenId / maximum
      Previous value: -48990New value: +29150
  2. First observedv0.8.0

TDQS

A5/5.0
Behavior5/5

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

No annotations are provided, so the description carries the full disclosure burden. It does so thoroughly: it warns about irreversible commitment, switch costs, the occupied process slot, early termination on dry deposits, no overrun beyond scheduled batches, and the fact the result reports what actually burned. This is exactly the kind of behavioral context an agent needs before mutating state.

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?

The description is long, but every sentence carries a distinct operational constraint or pointer that affects correct invocation. It is front-loaded with the core action and prerequisites, then moves from parameter semantics to lifecycle consequences, and includes concrete sibling-tool references. Nothing feels redundant or decorative.

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?

For a state-changing mining tool with no annotations and no output schema, the description is remarkably complete. It covers prerequisites, parameter selection rules, side effects, scheduling economics, cancellation prohibition, process-slot locking, and post-job claim/tracking steps. An agent has everything needed to decide whether, when, and how to invoke it.

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

Parameters5/5

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

Although the schema already documents all three parameters, the description adds meaning the schema cannot: targetResourceId behavior around switch cost and free first pick, batches being an irrevocable commitment, and tokenId requiring an owned cell with a finished extractor. It also explains the economic tradeoff of scheduling past deposit exhaustion, which materially changes how an agent should set batches.

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

Purpose5/5

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

The opening sentence states a specific verb ('start'), a precise resource ('a bounded extraction job'), and the required preconditions ('a cell you own that holds a finished extractor'). It clearly distinguishes this from siblings like cpu_build, cpu_get_mining_status, and cpu_claim_mining by framing it as the job-starting action.

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

Usage Guidelines5/5

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

The description explicitly says when to use the tool: after cpu_authenticate, on a finished extractor, with a valid mineable resource, and after checking the price in cpu_get_cell. It also tells the agent to size batches against game config and deposit, and points to cpu_get_mining_status and cpu_claim_mining for the subsequent lifecycle, so no invocation context is left to inference.

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

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/projectcpu/project-cpu-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server