Skip to main content
Glama

Run a board's checks

run_checks

Use when get_checks reports the checks have not been run for this board's current version and you need a verdict before answering. Runs KiCad's own DRC and ERC, or a fab house's manufacturability rules when vendor is given (list them with list_fab_profiles). EXPENSIVE and shared: this is a real kicad-cli run on one worker that everyone's boards queue behind, so run it when the answer matters, not on every board you look at. It usually returns the result directly. Status 'running' means a relevant check is queued or processing; read its result later with get_checks. Status 'busy' means another job holds the board's worker slot and nothing was started: call run_checks once more after retryAfterMs. Do not use it to re-run a board that already has a result at this version.

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": {
      "board": "<board>"
    },
    "name": "run_checks"
  }
}

Input Schema

NameRequiredDescriptionDefault
boardYesA personal board, or an organisation board with read:org consent and check permission: handle/slug or a boardrepo.com URL. Discover it with list_my_boards; do not guess one.
rulesNoOptional: YOUR OWN limits in millimetres, when no published fab profile fits - an in-house process, a tier a vendor does not publish, or a house rule tighter than any of them. Keys are the same limit names list_fab_profiles reports (trackWidthOuterMm, clearanceOuterMm, drillPlatedMinMm, annularMinMm and so on); values are millimetres. Mutually exclusive with vendor. The result is never stored as if a fab had published it.
vendorNoOptional fab house. Known ids: jlcpcb, pcbway, oshpark, multi-cb, eurocircuits, advanced-circuits, seeed, elecrow. A house name or one of that house's profile ids also resolves. Omit to run the board's own DRC + ERC.
versionNoWhich saved version to read: a version number (3 or "v3"), a versionId from list_board_versions, "latest", or a release tag such as "rev-A". Omit for the latest version. Results echo the version they read; pass its id back to stay on it.

Output Schema

NameRequiredDescriptionDefault

No arguments

Schema Changelog

Changes observed during successful MCP inspections.

  1. Added
  2. Removed
  3. Changed1 schema field changed
    • changedInput schema / properties / board / description
      Previous value: -"A public board: handle/slug (e.g. alice/keezyboost40) or a boardrepo.com URL. Discover it with search_boards; do not invent one."New value: +"A board: handle/slug or a boardrepo.com URL. Use search_boards for public boards or list_my_boards for personal and authorised organisation boards. Do not invent references."
  4. First observed

TDQS

A4.6/5.0
Behavior5/5

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

Goes well beyond the annotations by disclosing cost ('EXPENSIVE and shared... everyone's boards queue behind'), the shared-worker contention model, the meaning of each returned status ('running' vs 'busy' with retryAfterMs), and the guidance to poll with get_checks. None of that is derivable from readOnlyHint=false / openWorldHint=true / destructiveHint=false.

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?

Front-loaded with the trigger condition and the operation, then cost, then return/status handling, then the negative constraint — a sensible ordering with no filler sentences. It is dense but slightly long, and 'It usually returns the result directly' is hedged rather than precise.

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?

An output schema exists, so return shape needn't be described, and the description still covers the things structure cannot: cost, worker contention, status semantics, retry behavior, and permission expectations (read:org consent / check permission in the board param). An agent has everything needed to call it correctly and to react to its outcomes.

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 coverage is 100% and the schema already explains board, rules, vendor, and version in depth, so the description adds little parameter-level information. Its only contribution is pointing at list_fab_profiles for vendor ids and framing board as 'this board's current version', which is marginal over the schema text. Baseline 3 applies.

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?

States a specific verb and resource ('Runs KiCad's own DRC and ERC, or a fab house's manufacturability rules when `vendor` is given'), including the two operating modes and the profile source. An agent can immediately distinguish it from get_checks, which only reads existing results.

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?

Gives an explicit trigger ('when get_checks reports the checks have not been run for this board's current version'), an explicit exclusion ('Do not use it to re-run a board that already has a result at this version'), and names the sibling to consult for vendor ids (list_fab_profiles). The cost-based throttle ('run it when the answer matters, not on every board you look at') further bounds usage.

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

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources