Skip to main content
Glama
Sign In
Kiln

Kiln

by codeofaxel
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python

apply_calibration_recommendations

Destructive

Persists calibration tuning overrides as soft defaults, letting you save noise floor, confidence thresholds, and auto-refresh settings for future runs.

Instructions

Persist calibration tuning overrides as soft defaults.

Needs Kiln Pro.

Tool call template

This tool declares no required parameters. Add optional parameters from the input schema as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {},
    "name": "apply_calibration_recommendations"
  }
}

Input Schema

NameRequiredDescriptionDefault
sourceNoAudit attribution. One of: ``"user_explicit"`` (the user said apply), ``"agent_recommendation"`` (an LLM agent applied on the user's behalf), ``"auto_refreshed"`` (the recommendations engine self-refreshed — library-internal, callers should not pass this). Default: "user_explicit".
annotationNoFree-form context. Default: "".
noise_floor_mmNoThe smallest difference, in mm, between two readings that counts as a real disagreement rather than noise. ``-1.0`` (the default) leaves the current value unchanged. Default: -1.0.
based_on_sample_countNoAudit field — how many observations the recommendation was based on. Default: 0.
clarification_thresholdNoConfidence (0-1) below which a natural-language fit report is answered with ``clarification_needed`` instead of being recorded. ``-1.0`` (the default) leaves the current value unchanged. Default: -1.0.
auto_refresh_when_stableNoIf True, the recommendations engine will refresh this file automatically when its data-driven verdict has been stable across ``auto_refresh_stability_window`` consecutive recomputations. Default: false.
heuristic_confidence_levelsNoNew defaults for any subset of the four NL heuristic categories (``explicit_magnitude`` / ``matched_direction`` / ``hedged_direction`` / ``no_signal``). Missing keys keep the module default.
auto_refresh_stability_windowNoHow many consecutive stable recommendations trigger an auto-refresh. Default 3. Default: 3.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed8 schema fields changedv1.5.1
    • changedInput schema / properties / annotation / description
      Previous value: -"Default: \"\"."New value: +"Free-form context. Default: \"\"."
    • changedInput schema / properties / auto_refresh_stability_window / description
      Previous value: -"Default: 3."New value: +"How many consecutive stable recommendations trigger an auto-refresh. Default 3. Default: 3."
    • changedInput schema / properties / auto_refresh_when_stable / description
      Previous value: -"Default: false."New value: +"If True, the recommendations engine will refresh this file automatically when its data-driven verdict has been stable across ``auto_refresh_stability_window`` consecutive recomputations. Default: false."
    • changedInput schema / properties / based_on_sample_count / description
      Previous value: -"Default: 0."New value: +"Audit field — how many observations the recommendation was based on. Default: 0."
    • changedInput schema / properties / clarification_threshold / description
      Previous value: -"Default: -1.0."New value: +"Confidence (0-1) below which a natural-language fit report is answered with ``clarification_needed`` instead of being recorded. ``-1.0`` (the default) leaves the current value unchanged. Default: -1.0."
    • addedInput schema / properties / heuristic_confidence_levels / description
      Added value: +"New defaults for any subset of the four NL heuristic categories (``explicit_magnitude`` / ``matched_direction`` / ``hedged_direction`` / ``no_signal``).  Missing keys keep the module default."
    • changedInput schema / properties / noise_floor_mm / description
      Previous value: -"Default: -1.0."New value: +"The smallest difference, in mm, between two readings that counts as a real disagreement rather than noise.  ``-1.0`` (the default) leaves the current value unchanged. Default: -1.0."
    • changedInput schema / properties / source / description
      Previous value: -"Default: \"user_explicit\"."New value: +"Audit attribution.  One of: ``\"user_explicit\"`` (the user said apply), ``\"agent_recommendation\"`` (an LLM agent applied on the user's behalf), ``\"auto_refreshed\"`` (the recommendations engine self-refreshed — library-internal, callers should not pass this). Default: \"user_explicit\"."
  2. Addedv1.5.0

TDQS

B3.3/5.0
Behavior3/5

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

Annotations already flag destructiveHint=true and readOnlyHint=false, so the safety profile is known; the description adds that the persisted values are "soft defaults" and that Kiln Pro entitlement is required. It never explains what "soft default" means operationally, whether prior values are overwritten, or how to undo the change.

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?

Two short sentences, front-loaded with the action and followed by the prerequisite; there is no filler. It is efficient but borders on under-specification for a destructive tool.

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

Completeness2/5

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

For an 8-parameter destructive mutation with a nested object and no output schema, the description leaves the core mechanics unexplained: what exactly gets overwritten, whether existing calibration is merged or replaced, and what happens when the Kiln Pro check fails. The phrase "soft defaults" hints at reversibility but does not state it.

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% and each of the 8 parameters (including the -1.0 sentinel behavior and the heuristic_confidence_levels subset semantics) is documented in the schema. The description contributes no additional parameter meaning, so the baseline of 3 applies.

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?

"Persist calibration tuning overrides as soft defaults" names a specific verb (persist) and resource (calibration tuning overrides), so the effect is clear. It does not, however, distinguish itself from the obvious sibling revert_calibration_recommendations or from get_calibration_health_recommendations, which presumably produces the input for this tool.

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?

"Needs Kiln Pro" gives one hard prerequisite, which is genuinely useful for gating. But there is no explicit when-to-use/when-not guidance and no named alternative, so the agent must infer that this is the write-side counterpart to the recommendation/health tools.

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

Deploy Server

Other Tools