3D New Era AI

Overview Schema Source Related Servers Discussions Admin

Report to the developers

feedback

Report what a tool could do better with structured details (goal, tried, got, expected, cost, would_help, must_keep) to help developers make precise fixes.

Instructions

Report to the developers what a tool could do better, as it happens: a reply that answered less than was asked, a detour, a wrong conclusion reached before the right one, a fix that would distort the plan, something missing. It is a report the fix will be made from, so give the whole case — a vague note gets fixed in a way that breaks what already works. Every part is required: goal (the task and the situation: pieces, rooms, numbers), tried (the exact call with arguments), got (the literal reply), expected (what was true and how you found out), cost (the workaround, extra calls, wrong conclusion), would_help (the concrete change), must_keep (what works today and must not get worse with that change; "none known" is an answer). kind friction|bug|idea, tool. Kept on this machine and, while the user's telemetry is on, sent to the project. Reply {kept, sent}; then carry on with the task.

Input Schema

TableJSON Schema

Name	Required	Description
`got`	No	What came back, literally — the part of the reply that is the proof
`cost`	No	What it cost: the workaround, the extra calls, the wrong conclusion reached before the right one, the distortion made to the plan
`goal`	No	What you were trying to do, and the situation in the plan: the pieces, rooms and numbers involved
`kind`	No	`friction` (it worked, at a cost), `bug` (it answered wrong) or `idea` (something missing)
`tool`	No	The tool it is about, e.g. `measure`
`tried`	No	The exact call, with its arguments
`expected`	No	What was true instead, and how you found out (another tool, a render, a probe) — the reference the fix will be checked against
`must_keep`	No	What works today and must keep working with that change — the cases where the current behaviour is right. "none known" is an answer
`would_help`	No	The change that would have shortened the way, as concretely as you can: a field, a parameter, a rule

Output Schema

TableJSON Schema

Name	Required	Description	Default
`kept`	Yes
`sent`	Yes

Schema Changelog

Changes observed during successful MCP inspections.

Changed10 schema fields changedv3.0.0Sep 30, 2026
- removedInput schema / properties / cost / default
```
Removed value: -""
```
- removedInput schema / properties / expected / default
```
Removed value: -""
```
- removedInput schema / properties / goal / default
```
Removed value: -""
```
- removedInput schema / properties / got / default
```
Removed value: -""
```
- addedInput schema / properties / kind / enum
```
Added value: +[
+  "friction",
+  "bug",
+  "idea"
+]
```
- removedInput schema / properties / must_keep / default
```
Removed value: -""
```
- removedInput schema / properties / tried / default
```
Removed value: -""
```
- removedInput schema / properties / would_help / default
```
Removed value: -""
```
- removedOutput schema / properties / kept / description
```
Removed value: -"Where the report was written"
```
- removedOutput schema / properties / sent / description
```
Removed value: -"Whether it reached the developers"
```
First observedv0.1.0Sep 24, 2026

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 data handling ('Kept on this machine and, while the user's telemetry is on, sent to the project') and the response shape ('Reply {kept, sent}'). It also reveals a schema-invisible behavioral rule: although required parameters is 0, 'Every part is required,' and it warns that a vague note gets fixed in a way that breaks working behavior.

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 core action, and the dense paragraph largely earns its length by listing triggers, the completeness rationale, the required-field enumeration, kind/tool, and data handling. Some flourishes ('the whole case', 'shortened the way') add rhetorical weight without much information, so it is not maximally tight.

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?

For a 9-parameter reporting tool with an output schema, the description covers everything an agent needs: purpose, triggers, required fields, the kind enum's meaning, the destination of the data, and the return values. Nothing material is left undocumented.

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 genuine meaning beyond the schema: it asserts all nine fields are required despite required=0, and clarifies intent for ambiguous fields ('must_keep ... "none known" is an answer', 'expected ... the reference the fix will be checked against'). It mostly restates field intent rather than adding new syntax or format rules.

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?

The description states a specific verb and resource ('Report to the developers what a tool could do better') and is unmistakable against every sibling, which are all domain tools (measure, render_3d, edit_walls). An agent can tell instantly this is the meta-channel for filing a structured complaint about another tool.

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

Usage Guidelines4/5

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

It names concrete triggers for use — 'a reply that answered less than was asked, a detour, a wrong conclusion reached before the right one, a fix that would distort the plan, something missing' — and ties it to the moment it happens ('as it happens'). There is no explicit when-not-to-use, but no sibling offers overlapping function, so alternatives are not a real gap.

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