fix(agent): show skill name in permission prompt

[jonathanlab]

Problem

Invoking a skill produced a permission prompt that read just "Skill / Do you want to proceed?" — with no indication of which skill was being run.

The raw Claude SDK tool name Skill had no case in toolInfoFromToolUse (packages/agent/src/adapters/claude/conversion/tool-use-to-acp.ts), so it fell through to the default branch → title: "Skill", kind: "other". In the UI, kind: "other" routes to DefaultPermission, which renders the bare title. (Notably, ToolCallView already has a Skill display entry for the in-conversation view — only the permission path was missing it.)

Fix

Add a Skill case that surfaces the skill name from input.skill, so the prompt now reads "Skill: deep-research / Do you want to proceed?", and passes any args through as tool-call content.

Kept kind: "other" so it continues routing to DefaultPermission — no new component needed. A dedicated SkillPermission renderer (icon + styling) could follow later.

Verification

• pnpm --filter @posthog/agent typecheck → clean
• biome lint on the changed file → no issues

[trunk-io]

Merging to main in this repository is managed by Trunk.

• To merge this pull request, check the box to the left or comment /trunk merge below.

After your PR is submitted to the merge queue, this comment will be automatically updated with its status. If the PR fails, failure details will also be posted here

[github-actions]

React Doctor found no issues in the changed files. 🎉

_{Reviewed by React Doctor for commit d92b5d4.}

[greptile-apps]

_{Reviews (1): Last reviewed commit: "fix(agent): show skill name in permissio..." | Re-trigger Greptile}