Subprompt - AI Prompt Camera

Snap a photo of your meal and let any AI break down the macros. Shoot a receipt and get it itemized. Point your camera at a plant, a part, or a wine label and get an answer back. Subprompt does this without you typing the same instruction every time.

Here is the trick: every label you create is a saved AI prompt. When you capture, Subprompt writes that prompt right into the file. It stamps it onto the photo, sets it into a note's frontmatter, or speaks it at the start of a voice recording. Send the file to ChatGPT, Claude, Gemini, or any AI that reads images, text, or audio, and it already knows what to do. No copy-pasting prompts, no separate metadata files, no API keys.

Every capture saves to ANY folder on your phone, including a synced folder like an Obsidian vault, a Syncthing share, or Google Drive. Point Subprompt at a folder your desktop already watches and your captures flow straight into your AI pipelines and scheduled tasks, picked up and processed hands-free while you keep capturing.

THREE CAPTURE MODES, ONE WORKFLOW

📷 Image: take a photo, optionally type a note. Subprompt bakes a caption strip onto the bottom of the image with the label, prompt, note, and timestamp. Any vision-capable AI reads the prompt for free.

📝 Text: type a note. Saves as a markdown file with YAML frontmatter (label, prompt, captured time) and a body. Drop it into ChatGPT or Claude and the model already knows the job.

🎙️ Audio: record a voice note. On save, Subprompt prepends a spoken version of the prompt to the front of the recording, so it plays as prompt narration, then your voice. Feed it to any voice AI and the instruction is heard first.

LABELS ARE PROMPTS

Each label has a name, color, AI instruction, and a save folder. Image labels bake the prompt into the caption strip. Text labels write it into the markdown. Audio labels speak it at the start of every recording. Browse the built-in PROMPT LIBRARY for 40 ready-made templates across Food, Health, Finance, Productivity, Identification, Home, Pets, Travel, Learning, Hobbies, and Vehicles. Tap one to add it as a label.

FREE

- Up to 4 labels per mode (4 image, 4 text, 4 audio)

- Save to ANY folder on your phone. Most apps trap files in DCIM. Subprompt uses Android's Storage Access Framework so you choose exactly where every file lands

- Audio mode with live waveform, listen-before-send review, and a spoken prompt at the start of every recording

- 40-template Prompt Library

- Saved-files log of everything you have captured, filterable by type, label, and date

- Caption strip with label, prompt, note, and timestamp

- Tap-to-focus, pinch-to-zoom, flash, front and back camera

- Per-type preview viewer for images, audio, and markdown

PRO (one-time purchase) adds

- Unlimited labels per mode

- Sub-labels: multi-select tags baked into the image, spoken in the audio, or written into the markdown

- Auto-open the share sheet after save

- Embed location (city, country) into captures

- Remove the "Powered by Subprompt" branding from saved files

WHO IT IS FOR

- Diet logging: let an AI estimate macros from a meal photo

- Voice journaling: speak a thought, get a daily summary back

- Quick text notes with the prompt already attached for AI processing

- Receipts, plant ID, parts ID, supplement tracking, pet care

- Anyone syncing captures into an Obsidian vault or shared folder that a desktop AI pipeline watches and processes automatically

- Anyone tired of typing the same prompt twice

PRIVACY

Photos, notes, and recordings stay on your device. The only thing that ever leaves is your Play Store purchase, handled by Google and our billing provider. Crash reports are anonymous, with no email, no IP, and no personal data. No analytics, no ads, no accounts, no logins. Full privacy policy: https://nunncurtislabs.com/subprompt-privacy-policy/

Built by NunnCurtis Labs.