feat(code): show more context for MCP permission requests

[adboio]

Problem

MCP permissions requests lack context - e.g. here i have no idea if i want to approve exec lol

Changes

adds more context to the MCP permission requests

How did you test this?

manually / storybook

Publish to changelog?

no

[adboio]

Warning

This pull request is not mergeable via GitHub because a downstack PR is open. Once all requirements are satisfied, merge this PR as a stack on Graphite.
Learn more

• feat(code): show more context for MCP permission requests #2071 👈 (View in Graphite)
• feat(code): clean up plan question response key handling and validation #2070
• fix(code): fix styling on inline file links #2069
• feat(code): add diff stats to convo view #2065
• main

This stack of pull requests is managed by Graphite. Learn more about stacking.

[greptile-apps]

Comments Outside Diff (1)

1. apps/code/src/renderer/components/permissions/McpPermission.tsx, line 40 (link)

   The posthogDisplay.input field is designed as "Args to show as the input preview" (per the JSDoc in posthog-exec-display.ts), but fullInput always uses formatInput(toolCall.rawInput) — which for a call execute-sql {"query":"select 1"} invocation shows the outer {"command":"call execute-sql {…}"} wrapper instead of the cleaner extracted {"query":"select 1"}. The PostHogExecDisplay.input value goes entirely unused.

   Prompt To Fix With AI

   This is a comment left during a code review.
   Path: apps/code/src/renderer/components/permissions/McpPermission.tsx
   Line: 40
   
   Comment:
   The `posthogDisplay.input` field is designed as "Args to show as the input preview" (per the JSDoc in `posthog-exec-display.ts`), but `fullInput` always uses `formatInput(toolCall.rawInput)` — which for a `call execute-sql {"query":"select 1"}` invocation shows the outer `{"command":"call execute-sql {…}"}` wrapper instead of the cleaner extracted `{"query":"select 1"}`. The `PostHogExecDisplay.input` value goes entirely unused.
   
   
   
   How can I resolve this? If you propose a fix, please make it concise.

Prompt To Fix All With AI

Fix the following 1 code review issue. Work through them one at a time, proposing concise fixes.

---

### Issue 1 of 1
apps/code/src/renderer/components/permissions/McpPermission.tsx:40
The `posthogDisplay.input` field is designed as "Args to show as the input preview" (per the JSDoc in `posthog-exec-display.ts`), but `fullInput` always uses `formatInput(toolCall.rawInput)` — which for a `call execute-sql {"query":"select 1"}` invocation shows the outer `{"command":"call execute-sql {…}"}` wrapper instead of the cleaner extracted `{"query":"select 1"}`. The `PostHogExecDisplay.input` value goes entirely unused.

```suggestion
  const fullInput = posthogDisplay?.input ?? formatInput(toolCall.rawInput);
```

_{Reviews (1): Last reviewed commit: "feat(code): show more context for MCP pe..." | Re-trigger Greptile}