Coderbuds

Record Change Fit Decision Tool

record-change-fit-decision-tool

Idempotent

Say what you did about a assess-change-fit verdict. Call this right after you act on one — especially when you override it.

decision is one of: proceeded (opened it anyway), split (broke it into smaller PRs), trimmed (cut it down to fit), abandoned (dropped the change). Give a reason whenever you proceed: "the severable piece is dead code the change deletes — splitting would make review harder" is the kind of thing the team should be able to read back.

Overriding is a legitimate move and this is how you make it one. An unrecorded override is indistinguishable from never having got round to it, so the standard can only ever be obeyed, never argued with. Recorded overrides are what let Coderbuds notice a threshold that fires too often on changes that turn out fine, and move it. When that happens this response carries a standard_adjustment — say that once to the person, because the standard they are held to just changed, then let it drop.

Once you have recorded it, stop telling the person about it. The record is the report — it is what the team reads back — and repeating the verdict, the line count and the standard in your summary is the same argument made twice to someone who already agreed. A standard that has to be re-explained on every change it fires on is not being upheld, it is being nagged about, and people turn those off.

Defaults to the team's most recent undecided check, so you can usually call it with just decision and reason. Pass check_id (returned by assess-change-fit) to be explicit.

Input Schema

TableJSON Schema

Name	Required	Description
`reason`	No	Why — required in spirit when proceeding. Max 500 characters, visible to the team.
`check_id`	No	The check this answers (from assess-change-fit). Defaults to your most recent undecided check.
`decision`	Yes	What you did: proceeded (opened it anyway), split, trimmed, or abandoned.
`repository`	No	Optional "owner/name" repository slug, used only to bind the session to the owning team.

Output Schema

TableJSON Schema

Name	Required	Description
`team`	Yes	Team name.
`message`	Yes	Human-readable confirmation.
`verdict`	Yes	The verdict being decided on.
`check_id`	Yes	The fit check the decision was recorded against.
`decision`	Yes	The decision recorded.
`assessed_at`	Yes	When that check ran, ISO 8601.
`decision_reason`	No	The reason recorded, or null.
`standard_adjustment`	No	Present, once, when this override moved the team size standard: adjustment_id, previous_target_lines, target_lines, evidence, message.
`proposed_total_changes`	Yes	Raw total changed lines that were assessed.
`replaced_earlier_decision`	Yes	True when this overwrote a decision already on the check.

Schema Changelog

Changes observed during successful MCP inspections.

First observedSep 11, 2026

TDQS

A4.7/5.0

Behavior5/5

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

Annotations carry only idempotentHint, so the description does the heavy lifting: it discloses the default check binding, that the response may carry a `standard_adjustment`, that `reason` is visible to the team, and that the record is permanent and read back by the team. It also sets expectations about what the agent should and should not do after recording.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness3/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loaded with the purpose and the key instruction, but the description runs to six paragraphs with repeated persuasion (the "nagging" argument is made twice). The behavioral coaching earns some of its space, but the same points are restated rather than compressed.

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 values need not be enumerated, yet the description still flags the one non-obvious response field (`standard_adjustment`). Combined with defaults, override rationale, and post-call behavior, an agent has everything needed to invoke and follow through correctly.

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, but the description adds real meaning: plain-language glosses for each enum value in context, guidance that `reason` is "required in spirit" when proceeding, and how `check_id` defaults. It does not add format or length guidance beyond the schema's 500-char note.

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+resource: recording what you did about an `assess-change-fit` verdict. It explicitly names the sibling tool that produces the verdict it answers, so an agent can immediately place it in the workflow.

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 explicit timing ("call this right after you act on one — especially when you override it"), names the upstream alternative (assess-change-fit), and explains that overrides are a legitimate, expected case. The default behavior (most recent undecided check, so only `decision` and `reason` needed) is spelled out.

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

All Coderbuds tools Inspect

Try in Browser

Pay for results, not tokens.Fund a bounty. Let AI agents compete and collaborate to solve your problems.agentbounties.appAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources

Need Help?