TweakThis

List changes

list_changes

Read-onlyIdempotent

Lists what happened to the published feedback of one project since your last call, oldest first, at most 50 per call: new items, a developer's answer to your question, reopened items, items assigned to you and revised requests (and, when asked for in kinds, other status changes and comments). Each change has the item's id, a short title, its status, revision and version now, and who acted. Call it first without a cursor: it returns no changes and a cursor that starts from now. Then call it with the latest next_cursor instead of listing everything again, and at once again while has_more is true. Your own connection's comments, assignments and resolutions are left out unless include_own is true; claims are never listed. The cursor moves past the changes that kinds and session_id leave out, so keep the same filters for one cursor. Not a metered read; read an item with get_feedback before you act on it.

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": {
      "project_id": "<project_id>"
    },
    "name": "list_changes"
  }
}

Input Schema

Name	Required	Description
`kinds`	No	Which changes to return. Default: new, answered, reopened, assigned_to_me, revised.
`limit`	No	Changes per call (default and maximum 50).
`cursor`	No	next_cursor from your previous list_changes call on this project. Omitted: no changes, and a cursor from now.
`project_id`	Yes	Project id from list_projects.
`session_id`	No	Only items of this review session (the developer gives you its id).
`include_own`	No	Also your own connection's comments, assignments and resolutions. Default false.

Output Schema

Name	Required	Description
`error`	No	Present only when isError is true: a stable error code (not_found, invalid, conflict_stale, conflict_claimed, conflict_state, quota_exhausted, rate_limited, internal_error) with details such as current_revision/current_version.
`changes`	No
`has_more`	No
`next_cursor`	No
`schema_version`	Yes

Schema Changelog

Changes observed during successful MCP inspections.

AddedNov 7, 2026

TDQS

A4.8/5.0

Behavior5/5

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

Annotations already cover read-only, idempotent, closed-world and non-destructive, but the description adds substantial behavior beyond them: own connection's comments/assignments/resolutions are filtered out unless include_own is true, claims are never listed, and the cursor advances past changes excluded by kinds/session_id. It also discloses the billing profile ('Not a metered read') and the first-call empty-result semantics, which annotations cannot express.

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?

It is front-loaded with the purpose, then the paging workflow, then the filter caveats, so the most decision-relevant facts come first and every clause carries information. The single dense paragraph is close to run-on however, and a short bulleted split of the polling protocol would be more scannable.

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 cursor-paginated, filtered list tool this covers the full loop: first call, subsequent calls, termination condition, and filter stability. An output schema exists, yet the description still sketches the change payload (item id, title, status, revision, version, actor), so nothing an agent needs to call or interpret it is missing.

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 schema documents all six parameters and a baseline of 3 applies. The description still adds real meaning beyond it: the default kinds set, the interaction between cursor state and filters, and the fact that claims can never appear in results. That filter-consistency rule for cursor reuse is semantics the schema does not state, though the limit parameter gets no additional discussion.

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 ('Lists what happened to the published feedback of one project'), plus ordering (oldest first) and page size (at most 50). It enumerates the change kinds it surfaces and explicitly distinguishes itself from get_feedback ('read an item with get_feedback before you act on it'), so an agent can separate it from list_feedback and the write siblings without opening the schema.

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 bootstrap protocol ('Call it first without a cursor: it returns no changes and a cursor that starts from now'), an incremental-polling protocol ('call it with the latest next_cursor... at once again while has_more is true'), and a constraint ('keep the same filters for one cursor'). It also names the alternative for acting on a change (get_feedback) and the exclusion of claims, leaving little to inference.

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

All TweakThis tools Inspect

Try in Browser

Claude Code, inside the Godot 4 editorFlockbay puts an AI agent in the real Godot 4 editor: it writes GDScript, runs your game and fixes what breaks. Free, or bring your own Claude Code or Codex.flockbay.comAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources

Need Help?