You are helping me build the following AI agent workflow.

## Goal
IOS Meeting Notes to Obsidian via Local AI: Records an iOS voice memo, transcribes it locally, optionally summarizes with ChatGPT.

## Specification
- What it does: Capture your iOS voice memos and instantly turn them into organized Markdown notes stored in your Obsidian vault. The system transcribes audio locally on your device and optionally summarizes the content with ChatGPT before uploading everything to Google Drive for private, searchable access.
- Trigger: Run manually (Manual · on voice memo recording)
- Autonomy: Fully hands-off
- Expected setup effort: about an afternoon
- Tools/services involved:
  - iOS Voice Memos: Record audio input
  - n8n: Process webhook and convert to Markdown
  - Google Drive: Store and sync files
  - ChatGPT: Summarize transcription text

## Known pitfalls, handle each one explicitly in your implementation
1. Local transcription on iOS might fail for noisy audio, check the status code before sending to the LLM.
2. Markdown formatting errors can break Obsidian links, validate the output string structure before upload.

## Reference implementation
https://n8n.io/workflows/10482 (workflow template)
Fetch and inspect this before building. If it matches my stack, adapt it;
if not, rebuild the pattern with my tools.

## Process requirements
1. Before building: ask me which of the listed tools I actually use and
   what my platform is (n8n / Make / code / other). Do not assume.
2. Adapt the pattern to my answers; do not force the reference stack.
3. Address every pitfall above; tell me how you handled each.
4. Provide a test plan I can run before letting this touch real data.
5. Ask before any step that sends messages, modifies data, or spends money.

Source: https://usecasesforagents.com/use-case/ios-meeting-notes-to-obsidian-via-local-ai/ via usecasesforagents.com