Skip to main content
Glama

gate_check

Judge whether a proposal candidate may be adopted by checking anchors, parent base, diff safety, arms, budgets, guards, and noise; record promote, adopt-no-harm, or reject without changing the base.

Instructions

Judge whether a proposal's candidate may be adopted and record the verdict. It checks frozen anchors, that the proposal's parent is still the current base, an editable-only diff, a scan for task-specific literals, complete non-smoke selection arms of parent and candidate, cost and latency budgets, scenario quality guards and the recorded noise threshold. Use it after compare_paired looks promising; use ledger_adopt (operator profile) to act on a passing verdict, because this tool never changes the current base. Appends one verdict record per call. Returns {id, verdict, reasons, threshold, comparison, secondary_guards} with verdict promote (net above noise), adopt-no-harm (net >= 0 with a rationale) or reject.

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": {
      "config": "<config>",
      "proposal_id": "<proposal_id>",
      "parent_run": "<parent_run>",
      "child_run": "<child_run>",
      "noise_id": "<noise_id>",
      "aggregation": "<aggregation>"
    },
    "name": "gate_check"
  }
}

Input Schema

NameRequiredDescriptionDefault
configYesPath of the scenario YAML file on the server host, absolute or relative to the server's working directory, for example `scenario.yaml`. It selects the workspace; a server started with --config rejects any other path.
genericNoAttest that the mechanism generalizes beyond the observed tasks. Any passing verdict requires true. Default false.
noise_idYesnoise_id from noise_estimate on this parent in this epoch.
child_runYesRun id of a complete, non-smoke selection arm of the proposal's candidate, on the same tasks and repeats.
rationaleNoWhy the change is generic and safe. Required for an adopt-no-harm verdict. Default empty.
parent_runYesRun id of a complete, non-smoke selection arm of the proposal's parent.
aggregationYesHow repeats combine per task: `majority` for pass/fail outcomes (ties count as failure) or `mean` for numeric scores.
proposal_idYesProposal id from proposal_create for the candidate.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed10 schema fields changedv0.1.9
    • addedInput schema / properties / aggregation / description
      Added value: +"How repeats combine per task: `majority` for pass/fail outcomes (ties count as failure) or `mean` for numeric scores."
    • addedInput schema / properties / child_run / description
      Added value: +"Run id of a complete, non-smoke selection arm of the proposal's candidate, on the same tasks and repeats."
    • addedInput schema / properties / config / description
      Added value: +"Path of the scenario YAML file on the server host, absolute or relative to the server's working directory, for example `scenario.yaml`. It selects the workspace; a server started with --config rejects any other path."
    • addedInput schema / properties / generic / default
      Added value: +false
    • addedInput schema / properties / generic / description
      Added value: +"Attest that the mechanism generalizes beyond the observed tasks. Any passing verdict requires true. Default false."
    • addedInput schema / properties / noise_id / description
      Added value: +"noise_id from noise_estimate on this parent in this epoch."
    • addedInput schema / properties / parent_run / description
      Added value: +"Run id of a complete, non-smoke selection arm of the proposal's parent."
    • addedInput schema / properties / proposal_id / description
      Added value: +"Proposal id from proposal_create for the candidate."
    • addedInput schema / properties / rationale / default
      Added value: +""
    • addedInput schema / properties / rationale / description
      Added value: +"Why the change is generic and safe. Required for an adopt-no-harm verdict. Default empty."
  2. First observedv0.1.7

TDQS

A4.6/5.0
Behavior4/5

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

Adds real context beyond the annotations: it appends one verdict record per call (explaining the non-idempotent, non-read-only profile), never mutates the current base, and describes the three possible verdict values and their meanings. The only gap is that it does not spell out failure behavior when a guard check fails versus a hard error.

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 purpose and the when-to-use routing before the long checklist of guards, which is dense but each item is a distinct precondition an agent should know. The guard enumeration is the longest section and could arguably be trimmed, but it is not filler.

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?

No output schema exists, and the description compensates by naming the returned fields {id, verdict, reasons, threshold, comparison, secondary_guards} and defining each verdict value. Combined with the required-parameter list and the stated side effect of appending a record, an agent has everything needed to invoke and interpret it.

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 coverage is 100%, so the baseline is 3; the description earns a point by explaining how parameters interact operationally (parent_run must still be the current base, child_run must cover the same tasks and repeats, rationale is what enables adopt-no-harm). It does not add format or syntax detail beyond the schema, but the cross-parameter constraints are genuinely clarifying.

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 ('Judge whether a proposal's candidate may be adopted and record the verdict') and then enumerates the concrete checks it performs (frozen anchors, parent-is-base, editable-only diff, literal scan, selection arms, budgets, guards, noise threshold). This clearly separates it from siblings like compare_paired and ledger_adopt.

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 ('Use it after compare_paired looks promising') and names the alternative for the follow-up action ('use ledger_adopt ... to act on a passing verdict'), plus the key exclusion that this tool never changes the current base. When-to-use, when-to-escalate, and what it does not do are all stated.

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