Tallybook

Overview Schema Source Related Servers Discussions Admin

changes

Read-onlyIdempotent

Track each sub-agent model switch, compare real runs before and after, and see a verdict: keep, watch, revert, or too early. Measured figures, not estimates.

Instructions

Every point where a sub-agent's model changed, with the real runs before and after compared and a verdict: keep, watch, revert, or too early. These are measured figures, not estimates.

Input Schema

TableJSON Schema

Name	Required	Description
`since`	No	window: 7d, 30d, 90d, all, or a YYYY-MM-DD start date; default 30d
`source`	No	restrict to one tool: "claude-code" or "codex"; default both
`project`	No	restrict to one project directory, exact match, or a prefix if it ends with a path separator
`currency`	No	override the detected plan: "usd" reports as charged, "share" reports as a subscription's list-price equivalent
`min_runs`	No	runs required on each side of a change before judging it, at least 1; default 3
`include_undecided`	No	also return changes with too few runs on one side to judge; default false

Output Schema

TableJSON Schema

Name	Required	Description
`plan`	Yes	"api" or "subscription", re-detected on every scan
`changes`	Yes
`currency`	Yes	"usd" when the figures were charged, "list_price_equivalent" when a subscription covered them and the figures are what the usage would have cost on the API
`min_runs`	Yes
`undecided`	Yes	changes with too few runs to judge, whether or not they are included
`scan_error`	No	why the most recent scan attempt failed, in which case the figures are from the last successful scan
`age_seconds`	Yes	seconds since that scan; the server rescans when a tool is called more than 60 seconds after the last attempt, or on refresh
`ingested_at`	Yes	RFC 3339 time of the last successful transcript scan; empty if none has succeeded yet
`plan_reason`	Yes	why that plan was chosen

Schema Changelog

Changes observed during successful MCP inspections.

First observedv0.1.0Sep 12, 2026

TDQS

B3.4/5.0

Behavior3/5

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

Annotations already declare readOnlyHint, idempotentHint, and a closed-world profile, so safety is covered structurally. The description adds useful behavioral context by defining the verdict vocabulary and asserting the figures are measured rather than estimated, but it says nothing about pagination, ordering, or how the comparison is scoped.

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, no filler, and the core purpose is front-loaded before the qualifier about measured figures. Every clause carries information.

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 a full input schema, an output schema, and annotations, the description does not need to describe returns or safety. It covers the tool's purpose and output semantics adequately, though a note on how verdicts are computed or when the tool is the right entry point would close the remaining gap.

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 documents all six parameters (since windows, source, project matching, currency, min_runs, include_undecided). The description adds no parameter-level meaning beyond that, which puts it at the baseline of 3.

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 names a specific resource (points where a sub-agent's model changed) and a concrete output (before/after runs compared with a keep/watch/revert/too early verdict), which is clearly distinct from siblings like agents, prices, or report. It stops short of explicitly contrasting itself with those siblings, so it is not a 5.

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

Usage Guidelines2/5

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

There is no guidance on when to reach for this tool versus agents, report, or session, and no mention of prerequisites or conditions. The closing line about measured figures implies a data-quality context but gives no when/when-not framing.

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

Deploy Server