[Improve] Streamline DCG command approval

[navedmerchant]

What changed

Reads the persisted DCG setting without eagerly building full provider state, restores full state loading until after approval, and adds coverage for an allowed command continuing through execution and for command rows awaiting a decision.

Why this change was made

The DCG integration should avoid unnecessary state work on ordinary commands while preserving explicit approval for guarded commands. Closes#1058 and completes #1049.

Impact

Command approval remains responsive when DCG is disabled, while guarded-command behavior and denied-state rendering are covered more completely.

Related PRs

• Managed binary infrastructure
• DCG binary service
• DCG setting
• DCG documentation

Summary by CodeRabbit

• New Features

  • Added destructive command protection with localized block messages, reasons, and rule details.
  • Commands approved by the protection system can bypass standard approval rules.
  • Added clear denied-command indicators in chat, with support for multiple languages.

• Bug Fixes

  • Commands denied by protection now require explicit approval and fail safely when protection is unavailable.
  • Hidden command-pattern controls prevent changes when protection is active or a command is denied.

• Tests

  • Expanded coverage for approval precedence, blocked commands, denied states, localization, and fail-closed behavior.

[coderabbitai]

📝 Walkthrough

Walkthrough

The change integrates destructive-command-guard evaluation with command auto-approval, persists approval decisions on messages, adds localized denial details, and updates chat rendering for denied commands.

Changes

Destructive command approval and persistence

Layer / File(s)	Summary
Persist auto-approval decisions packages/types/src/message.ts, packages/types/src/__tests__/message.test.ts, src/core/task/Task.ts, AGENTS.md	Messages now store "approve" or "deny" decisions. Task ask messages persist these decisions across completion paths.
Apply DCG approval precedence src/core/auto-approval/index.ts, src/core/auto-approval/__tests__/dcg.spec.ts	Protected commands require explicit approval. DCG-approved commands can bypass configured command lists when the required auto-approval settings are enabled.
Evaluate commands before execution src/core/tools/ExecuteCommandTool.ts, src/core/tools/__tests__/executeCommandTool.spec.ts, src/i18n/locales/*/tools.json	Commands are installed and evaluated by DCG before approval. Blocked commands include localized reason and rule details. Installation and unavailable-guard failures stop evaluation, approval, and execution.
Render denied command state webview-ui/src/components/chat/ChatRow.tsx, webview-ui/src/components/chat/CommandExecution.tsx, webview-ui/src/components/chat/__tests__/*, webview-ui/src/i18n/locales/*/chat.json	Denied commands display a localized status and denial icon. Pattern controls are hidden when DCG is enabled or the command is denied. Tests cover denied, approved, and undecided states.

Estimated code review effort: 4 (Complex) | ~45 minutes

Sequence Diagram(s)

sequenceDiagram
participant ExecuteCommandTool
participant DCG
participant AutoApproval
participant Task
participant ChatRow
ExecuteCommandTool->>DCG: Install and evaluate command
DCG-->>ExecuteCommandTool: Allow or denial details
ExecuteCommandTool->>AutoApproval: Request approval
AutoApproval-->>ExecuteCommandTool: Approve or deny
ExecuteCommandTool->>Task: Persist autoApprovalDecision
Task-->>ChatRow: Render command message state

Loading

Possibly related PRs

• Zoo-Code-Org/Zoo-Code#1050: Directly overlaps the DCG approval, command execution, persisted decision, and denied-command UI changes.
• Zoo-Code-Org/Zoo-Code#1060: Provides the DCG manager and runner integrated by ExecuteCommandTool.
• Zoo-Code-Org/Zoo-Code#1061: Adds the persisted destructiveCommandGuardEnabled setting consumed by this change.

Suggested labels:awaiting-review

Suggested reviewers:hannesrudolph

🚥 Pre-merge checks | ✅ 3 | ❌ 2

❌ Failed checks (2 warnings)

Check name	Status	Explanation	Resolution
Description check	⚠️ Warning	The description explains the change, rationale, impact, and issue links, but omits the required Test Procedure and Pre-Submission Checklist sections.	Add the template sections, document reproducible tests and environment details, and complete the checklist including documentation impact.
Docstring Coverage	⚠️ Warning	Docstring coverage is 0.00% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (3 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title clearly summarizes the main change: streamlining destructive command guard approval.
Linked Issues check	✅ Passed	The implementation covers DCG precedence, explicit approval, disabled behavior, denied persistence and rendering, localized details, hidden controls, and regression tests for [#1058].
Out of Scope Changes check	✅ Passed	All code and localization changes support DCG approval precedence, execution, denied-state UX, or regression coverage; no unrelated code changes are evident.

✨ Finishing Touches 💡 1

📝 Generate docstrings 💡

• Create stacked PR
• Commit on current branch

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch feat/dcg-command-integration

Warning

There were issues while running some tools. Please review the errors and either fix the tool's configuration or disable the tool if it's a critical failure.

🔧 ESLint

If the error stems from missing dependencies, add them to the package.json file. For unrecoverable errors (e.g., due to private dependencies), disable the tool in the CodeRabbit configuration.

packages/types/src/__tests__/message.test.ts

ESLint skipped: missing config or dependency (missing-dependency). The ESLint configuration references a package that is not available in the sandbox.

packages/types/src/message.ts

ESLint skipped: the ESLint configuration for this file references a package that is not available in the sandbox.

src/core/auto-approval/__tests__/dcg.spec.ts

ESLint skipped: missing config or dependency (missing-dependency). The ESLint configuration references a package that is not available in the sandbox.

• 8 others

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[codecov]

Codecov Report

❌ Patch coverage is 91.17647% with 3 lines in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
src/core/tools/ExecuteCommandTool.ts	90.47%	1 Missing and 1 partial ⚠️
src/core/task/Task.ts	50.00%	1 Missing ⚠️

📢 Thoughts on this report? Let us know!

[edelauna]

some minor questions

[zoomote]

Addressed the implementation and coverage feedback in 900282f. DCG reads the persisted flag cheaply before approval, preserves deferred full-state loading, and covers allow and pending chat-row paths.

[edelauna]

Thanks for addressing the feedback, approving - but still unclear on how you want to handle the design for a DCG blocked command: #1062 (comment)

The base branch was changed.

[coderabbitai]

🧹 Nitpick comments (2)

src/core/tools/__tests__/executeCommandTool.spec.ts (1)

243-247: 📐 Maintainability & Code Quality | 🔵 Trivial | ⚡ Quick win

Use typed test doubles instead of repeated double assertions.

Define typed Task and callback test doubles once in the fixture. Remove the repeated as unknown as casts. If a cast is unavoidable, document why near the fixture.

• src/core/tools/__tests__/executeCommandTool.spec.ts#L243-L247: pass typed Task and callback doubles.
• src/core/tools/__tests__/executeCommandTool.spec.ts#L266-L270: reuse the typed doubles.
• src/core/tools/__tests__/executeCommandTool.spec.ts#L284-L288: reuse the typed doubles.
• src/core/tools/__tests__/executeCommandTool.spec.ts#L304-L308: reuse the typed doubles.
• src/core/tools/__tests__/executeCommandTool.spec.ts#L329-L333: reuse the typed doubles.

As per coding guidelines: “Avoid as any; use typed APIs, bracket notation for private members, or precise test doubles and type guards. Use double assertions only as a last resort and explain them with a nearby comment.”

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
In `@src/core/tools/__tests__/executeCommandTool.spec.ts` around lines 243 - 247,
Replace the repeated double assertions in the executeCommandTool.handle test
calls with typed Task and callback test doubles defined once in the fixture.
Reuse those doubles at src/core/tools/__tests__/executeCommandTool.spec.ts lines
243-247, 266-270, 284-288, 304-308, and 329-333; if any cast remains necessary,
document its reason beside the fixture.

Source: Coding guidelines

webview-ui/src/components/chat/__tests__/CommandExecution.spec.tsx (1)

114-140: 📐 Maintainability & Code Quality | 🔵 Trivial | ⚡ Quick win

Type the extension-state fixtures instead of using as any.

mockExtensionState, ExtensionStateWrapper, and the local state passed to ExtensionStateContext.Provider are all cast to any, which removes compile-time checks for required context fields like destructiveCommandGuardEnabled, setAllowedCommands, and setDeniedCommands. Use the typed context value type or a typed wrapper in CommandExecution.spec.tsx.

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
In `@webview-ui/src/components/chat/__tests__/CommandExecution.spec.tsx` around
lines 114 - 140, Replace the any casts in the CommandExecution tests with the
actual ExtensionStateContext value type, including the local state and
ExtensionStateWrapper fixtures. Ensure mockExtensionState and the provider value
satisfy required fields such as destructiveCommandGuardEnabled,
setAllowedCommands, and setDeniedCommands without weakening type checking.

Source: Coding guidelines

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Nitpick comments:
In `@src/core/tools/__tests__/executeCommandTool.spec.ts`:
- Around line 243-247: Replace the repeated double assertions in the
executeCommandTool.handle test calls with typed Task and callback test doubles
defined once in the fixture. Reuse those doubles at
src/core/tools/__tests__/executeCommandTool.spec.ts lines 243-247, 266-270,
284-288, 304-308, and 329-333; if any cast remains necessary, document its
reason beside the fixture.
In `@webview-ui/src/components/chat/__tests__/CommandExecution.spec.tsx`:
- Around line 114-140: Replace the any casts in the CommandExecution tests with
the actual ExtensionStateContext value type, including the local state and
ExtensionStateWrapper fixtures. Ensure mockExtensionState and the provider value
satisfy required fields such as destructiveCommandGuardEnabled,
setAllowedCommands, and setDeniedCommands without weakening type checking.

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Pro Plus

Run ID: 84b524a3-87e4-4028-9574-e9338ea3e230

📥 Commits

Reviewing files that changed from the base of the PR and between 0ae9217 and 10476b3.

📒 Files selected for processing (48)

• AGENTS.md
• packages/types/src/__tests__/message.test.ts
• packages/types/src/message.ts
• src/core/auto-approval/__tests__/dcg.spec.ts
• src/core/auto-approval/index.ts
• src/core/task/Task.ts
• src/core/tools/ExecuteCommandTool.ts
• src/core/tools/__tests__/executeCommandTool.spec.ts
• src/i18n/locales/ca/tools.json
• src/i18n/locales/de/tools.json
• src/i18n/locales/en/tools.json
• src/i18n/locales/es/tools.json
• src/i18n/locales/fr/tools.json
• src/i18n/locales/hi/tools.json
• src/i18n/locales/id/tools.json
• src/i18n/locales/it/tools.json
• src/i18n/locales/ja/tools.json
• src/i18n/locales/ko/tools.json
• src/i18n/locales/nl/tools.json
• src/i18n/locales/pl/tools.json
• src/i18n/locales/pt-BR/tools.json
• src/i18n/locales/ru/tools.json
• src/i18n/locales/tr/tools.json
• src/i18n/locales/vi/tools.json
• src/i18n/locales/zh-CN/tools.json
• src/i18n/locales/zh-TW/tools.json
• webview-ui/src/components/chat/ChatRow.tsx
• webview-ui/src/components/chat/CommandExecution.tsx
• webview-ui/src/components/chat/__tests__/ChatRow.command-denied.spec.tsx
• webview-ui/src/components/chat/__tests__/CommandExecution.spec.tsx
• webview-ui/src/i18n/locales/ca/chat.json
• webview-ui/src/i18n/locales/de/chat.json
• webview-ui/src/i18n/locales/en/chat.json
• webview-ui/src/i18n/locales/es/chat.json
• webview-ui/src/i18n/locales/fr/chat.json
• webview-ui/src/i18n/locales/hi/chat.json
• webview-ui/src/i18n/locales/id/chat.json
• webview-ui/src/i18n/locales/it/chat.json
• webview-ui/src/i18n/locales/ja/chat.json
• webview-ui/src/i18n/locales/ko/chat.json
• webview-ui/src/i18n/locales/nl/chat.json
• webview-ui/src/i18n/locales/pl/chat.json
• webview-ui/src/i18n/locales/pt-BR/chat.json
• webview-ui/src/i18n/locales/ru/chat.json
• webview-ui/src/i18n/locales/tr/chat.json
• webview-ui/src/i18n/locales/vi/chat.json
• webview-ui/src/i18n/locales/zh-CN/chat.json
• webview-ui/src/i18n/locales/zh-TW/chat.json