enhancement: copy MCP permission

[urmauur]

Describe Your Changes

This pull request refines the tool approval dialog and updates localization strings across multiple languages to clarify the scope of tool permissions and improve consistency. The main focus is on making it clear that permissions granted for tools apply only to the current conversation/thread, not globally, and on improving the user interface for tool approval actions.

Tool Approval Dialog Improvements:

• Updated the tool approval dialog in ToolApproval.tsx to display a new description about the permission scope, and improved button styling and labeling for clarity. [1] [2]

Localization and Permission Scope Clarifications:

• Added or updated the permissionScope string in tools.json for all supported languages to clarify that granted permissions only apply to the current conversation/thread. [1] [2] [3] [4] [5] [6]
• Changed the label for "Always Allow" to "Allow in thread" (or equivalent) in tool approval dialogs and tool permission settings across all supported languages for consistency. [1] [2] [3] [4] [5] [6] [7]

Global Permission Setting Clarifications:

• Updated the allowPermissionsDesc string in mcp-servers.json for all supported languages to clarify that enabling global tool permissions applies to all conversations, including new chats. [1] [2] [3] [4] [5] [6]

Minor UI/UX Adjustments:

• Improved button capitalization and spacing in the tool approval dialog for better visual consistency.

These changes ensure users have a clearer understanding of how tool permissions are applied and improve the overall experience when approving tool actions.

Fixes Issues

• Closes #
• Closes #

Self Checklist

• Added relevant comments, esp in complex areas
• Updated docs (for bug fixes / features)
• Created issues for follow-up changes or refactoring needed

---

Important

Refines tool approval dialog and updates localization strings to clarify tool permission scope and improve consistency across languages.

• Tool Approval Dialog:
  • Updated ToolApproval.tsx to include a new description about permission scope and improved button styling.
• Localization Updates:
  • Added permissionScope string in tools.json for all supported languages to specify permissions apply only to the current conversation.
  • Changed "Always Allow" to "Allow in thread" in tool approval dialogs across all languages.
• Global Permission Clarification:
  • Updated allowPermissionsDesc in mcp-servers.json to clarify global tool permissions apply to all conversations, including new chats.

^{This description was created by for a26445e. You can customize this summary. It will automatically update as commits are pushed.}

[ellipsis-dev]

Caution

Changes requested ❌

Reviewed everything up to a26445e in 1 minute and 55 seconds. Click for details.

• Reviewed 325 lines of code in 21 files
• Skipped 0 files when reviewing.
• Skipped posting 6 draft comments. View those below.
• Modify your settings and rules to customize what types of comments Ellipsis leaves. And don't forget to react with 👍 or 👎 to teach Ellipsis.

1. web-app/src/containers/dialogs/ToolApproval.tsx:55

• Draft comment:
  Good enhancement: appending the permission scope clarification below the tool name improves clarity. Verify that punctuation and spacing are consistent.
• Reason this comment was not posted:
  Confidence changes required: 0% <= threshold 50% None

2. web-app/src/containers/dialogs/ToolApproval.tsx:101

• Draft comment:
  Nice addition of the 'capitalize' class on the allow button to improve UI consistency.
• Reason this comment was not posted:
  Confidence changes required: 0% <= threshold 50% None

3. web-app/src/locales/en/tool-approval.json:3

• Draft comment:
  Possible unintended 'hello' text in the description; consider removing it if it was left by mistake.
• Reason this comment was not posted:
  Comment did not seem useful. Confidence is useful = 0% <= threshold 50% The comment is asking the PR author to consider removing a text from the description, which is not allowed as per the rules. The rules explicitly state not to ask the PR author to update the PR description.

4. web-app/src/locales/en/mcp-servers.json:37

• Draft comment:
  The updated allowPermissionsDesc now clearly states that the setting applies globally to all conversations, including new chats. The revised text is clear and appropriate.
• Reason this comment was not posted:
  Confidence changes required: 0% <= threshold 50% None

5. web-app/src/locales/en/tools.json:10

• Draft comment:
  The addition of the new 'permissionScope' key helps clarify that permissions granted apply only to the current conversation. This consistency across locales is beneficial.
• Reason this comment was not posted:
  Confidence changes required: 0% <= threshold 50% None

6. web-app/src/locales/zh-CN/tools.json:7

• Draft comment:
  The translation for "alwaysAllow" was changed from "始终允许" to "在线程中允许". Please double-check if "在线程中允许" is intentional, as it might be a typographical error.
• Reason this comment was not posted:
  Decided after close inspection that this draft comment was likely wrong and/or not actionable: usefulness confidence = 10% vs. threshold = 50% This is a UI translation change. The new translation actually seems more accurate since line 10 explicitly states that permissions only apply to the current conversation. The old "always allow" was potentially misleading. We should trust that the PR author (likely a Chinese speaker) made this change intentionally to improve accuracy. I might be wrong about the nuances of Chinese translation. Maybe there's a better way to express this concept in Chinese. While translation nuances can be complex, our rules explicitly state not to comment on UI changes and to assume the author made UI changes correctly. This is clearly a UI text change. Delete the comment. This is a UI translation change that we should assume was made correctly by the author, and the new translation actually appears to better match the stated permission scope.

Workflow ID: wflow_t5nhdgce9N4Whvai

^{You can customize by changing your verbosity settings, reacting with 👍 or 👎, replying to comments, or adding code review rules.}

[github-actions]

Barecheck - Code coverage report

Total: 29.38%

Your code coverage diff: 0.00% ▴

Uncovered files and lines

File	Lines
web-app/src/containers/dialogs/ToolApproval.tsx	1, 9-12, 14-16, 18-20, 22, 24-26, 28-30, 32-34, 36-41, 43-62, 64-74, 77-81, 83-87, 89-95, 97-103, 105-110, 112