Documentation Index

Fetch the complete documentation index at: https://help.1nspect.app/llms.txt

Use this file to discover all available pages before exploring further.

​

Capture Screen

The Capture Screen is the heart of the 1nspecT mobile app. It’s the full-screen workspace where you actually capture findings — photos, captions, narratives, ratings — for every section of an inspection. You will spend most of your inspection time on this one screen. This article is the definitive reference for everything on the Capture Screen. If you’re new, run through the Mobile Inspection Walkthrough first for a guided practice flow, then come back here for the deeper detail.

---

​

When you see this screen

You reach the Capture Screen by tapping Start Inspection at the bottom of the Starting an Inspection setup form. The Capture Screen is presented as a full-screen modal — it covers the bottom tab bar so you can focus on the inspection without the tabs distracting. To leave the Capture Screen (intentionally), use the back arrow in the header. The app keeps your work safe — local data is autosaved as you go, so you can return at any time via Config → 🚀 Resume Active Inspection or via the In Progress card on Assigned Inspections.

---

​

Anatomy of the screen

[SCREENSHOT: capture-screen-full.png — mobile, the full Capture Screen showing all sections in order, top to bottom: photo preview area, capture controls, pending Co-Inspector pills, section nav, subsection nav, caption field, library chips, AI tools row, narrative field, field note, rating buttons, flag pills, save row.] Top to bottom:

1. Photo / video preview area — large area at the top showing the current finding’s media
2. Capture controls — three buttons: 📷 Photo · 🎥 Video · 📁 Library Import
3. Pending Co-Inspector pills (if any) — green chips for AI-suggested findings you queued for follow-up
4. Section navigation bar — horizontal scroll of report sections (e.g. Roofing, Structural, Electrical)
5. Subsection navigation bar — horizontal scroll of subsections within the active section
6. Caption field — short title for this finding (6–15 words)
7. FREE (Library) chips — auto-suggested narrative chips from your library as you type the caption
8. AI Tools row — ✦ AI Generate, 👁 Vision Scan, 🪄 Library, ℹ️ Sys Info
9. Narrative field — the full finding text (CAR format)
10. Field Note — private inspector notes, never shown in the report
11. Rating buttons — Inspected / Not Inspect. / Deficient / Not Present
12. Flag pills — 🔗 Link to master · ⚠️ Include in report summary
13. Save row — 💾 Save + New · ✔ Save + Finish · 🗑 Delete

Each section is covered in detail below.

---

​

1. Photo / video preview area

The preview area shows the currently attached media for the finding being captured.

​

States the preview shows

​

Controls on the preview area

When media is present, four secondary buttons appear over the preview: The Shrink/Expand toggle is genuinely useful — on a long capture session you can expand the preview to validate what you photographed, then shrink to get the form fields back. The setting persists per session. See Photos & Media for the full media capture workflow.

---

​

2. Capture controls

A row of three round buttons below the preview area:

Permissions: if you decline a permission the first time, the app shows a “Permission needed” alert each time and gracefully aborts the action. To grant later, open device Settings → 1nspecT → Camera / Microphone / Photos and enable.

After a successful photo capture, the app automatically downscales the image to ~1024 px wide at 0.7 JPEG quality and writes it to the device’s documents directory before adding it to the finding. The original-resolution shot is not kept — this is intentional, to keep the upload queue lean. If you need higher resolution for a specific finding (e.g. a serial-number photo), import from the device library instead.

---

​

3. Pending Co-Inspector pills

[SCREENSHOT: coinspector-pills.png — mobile, Capture Screen showing 2 green chips with the 📋 icon between the capture controls and the section nav bar. First chip caption: “GFCI missing at exterior outlet”. Second: “Worn fascia at front elevation”.] If you ran Vision Scan (the 👁 button) on a previous finding and the AI proposed additional deficiencies it spotted in the same photo, accepted ones appear here as green pills, each starting with a 📋 icon. Tap a pill to pre-populate the form with that pending finding’s caption and narrative, then capture a new photo of that specific deficiency, adjust the section/subsection, and save it. See AI During Capture → Co-Inspector for the full workflow.

---

​

4. Section navigation bar

The horizontal bar showing the report’s main sections (Roofing, Structural, Electrical, Plumbing, etc., depending on your template). The active section is highlighted.

• Tap a section to switch the active section. The subsection bar (below) updates to show that section’s subsections.
• Scroll horizontally if your template has more sections than fit on screen.

The section determines where the finding will appear in the final PDF report. Always confirm you’re on the right section before saving — moving a finding later (via the Review screen) is possible but adds steps.

---

​

5. Subsection navigation bar

Below the section bar, the horizontal bar of subsections within the active section. Examples within “Electrical”:

• Service Entry
• Main Panel
• Branch Circuits
• Outlets & Fixtures
• Smoke Detectors

Tap a subsection to select it. The current selection persists across saves until you change it — useful when capturing multiple findings within the same area.

Naming: the subsection name shown on this bar is the short name from the template schema (typically under 20 characters to fit). The full name appears on the PDF report.

---

​

6. Caption field

A short text input — 6–15 words is the sweet spot — describing the finding in inspector shorthand. Examples:

• Damaged shingles at rear slope — granule loss visible
• Double-tapped breaker at main panel
• Open ground at exterior GFCI outlet

The caption serves three purposes:

1. It’s the title of the deficiency card in the final PDF report.
2. It’s the input the AI uses for ✦ AI Generate and 🪄 Library Search. A specific caption produces a sharper AI narrative; a vague caption produces a vague one.
3. It triggers the FREE Library chips (see below) — as you type, the app keyword-matches your narrative library and offers matches.

The caption field accepts a “Search captions…” lookup if your inspection has a caption library assigned — useful for inspectors who reuse the same captions across jobs.

---

​

7. FREE (Library) chips

[SCREENSHOT: free-library-chips.png — mobile, just below the caption field, a row labeled “FREE (Library)” with 2 tappable green chips matching the typed caption.] As you type the caption, the app runs a local keyword match against your narrative library and offers matches as tappable chips, labeled FREE (Library). Tap a chip to insert that narrative into the Narrative field.

• Cost: 0 IT — completely free, runs entirely on-device
• Works offline — no network required
• Best for: common repeat findings where your library already has the language you need (roof shingles, GFCI outlets, double-tapped breakers, water heater age, etc.)

If no chips appear, your library has no keyword match yet. Use ✦ AI Generate or 🪄 Library Search (semantic) for AI-assisted alternatives.

---

​

8. AI Tools row

The row of AI action buttons. All cost IT tokens (except FREE Library chips above). The full reference is in AI During Capture; a quick map here: The Vision Scan toggle is off by default. Turn it on before tapping AI Generate to use the photo-grounded path. There’s a small NEW badge on Vision Scan because it’s the newest AI capability.

Error states you may see:

• “Photo Required — Vision Analysis requires a photo. Capture a photo first, or disable the Vision toggle…” — capture a photo before running Vision
• “Insufficient IT Tokens — Vision Analysis requires 8 IT…” — your tenant balance is too low; your administrator tops up on the web admin
• “⚠️ AI Offline — AI service unreachable. A template CAR narrative has been pre-filled…” — graceful fallback when the AI service is unreachable

---

​

9. Narrative field

The full inspector prose explaining the finding. This is the body text on the deficiency card in the final report. The narrative field supports:

• Manual typing of free-form text
• AI insertion — generated narratives populate this field directly
• Library insertion — tapping a FREE chip or a Library Search result fills this field
• In-field search — a “Search narratives…” affordance lets you find a narrative from your library without leaving the field

A small Save to Library action below the narrative (toast confirms after a brief delay) saves your current narrative back to the library for future reuse. Saving de-identifies the text (removes addresses, client names, etc.) before storage. Cost: 1 IT.

---

​

10. Field Note (optional)

A private, unformatted observation note that is never shown in the report. Placeholder text reads: “Optional: inspector context for AI (e.g. ‘not likely required at time of construction’) — never shown in report”. Field notes serve two purposes:

1. Personal context for yourself during long inspections — “this was a 1960s addition” — that you don’t want in the customer-facing report.
2. AI Draft fuel. If you skip writing the full narrative on-site and instead jot a field note, your office can use the AI Draft tool on the web admin to batch-generate full CAR narratives from your field notes after the fact. Costs 2 IT per drafted finding, runs from the Report Workspace.

If you write a full narrative on-site, you can skip the field note entirely.

---

​

11. Rating buttons — CONDITION RATING

A row of four rating buttons. Every required subsection must have a rating before the inspection can be finished.

The button label Not Inspect. is truncated to fit the row — the full rating is “Not Inspected”.

A subsection can carry both I Inspected AND D Deficient ratings simultaneously — for example, “I inspected the panel and noticed a deficiency in one breaker.” — which TREC permits. The mobile UI accepts this multi-rating model.

NI reminder: tapping NI surfaces a reminder toast prompting you to include the reason and any recommendations in the narrative field. Reports are stronger when an NI explains why the component couldn’t be inspected.

---

​

12. Flag pills

[SCREENSHOT: flag-pills.png — mobile, two small pill-shaped toggles below the rating row: 🔗 Link to master (chain icon) and ⚠️ Include in report summary (warning icon).] Two small toggleable pills below the rating row: Tap to toggle on/off. Tapping the same pill again turns it off.

---

​

13. Save row

[SCREENSHOT: save-row.png — mobile, three buttons at the bottom of the Capture Screen: 💾 Save + New (primary navy, fills most of the row), ✔ Save + Finish (neutral border), and 🗑 (red icon, square delete button on the far right).] Three buttons at the bottom of the screen: During a save, the active button briefly shows a spinner (“Saving…”); the other buttons are disabled to prevent double-tapping.

​

What a save records

When you tap Save + New or Save + Finish, the app commits:

• The current section and subsection
• The caption
• The narrative (and the field note, if you wrote one)
• All attached photos (uploads to cloud queue in background) and videos
• The rating (I / NI / NP / D)
• The flag pill states (link-to-master, include-in-summary)
• The system info tag, if you used Sys Info
• A createdAt timestamp + your inspector identity

All written to local storage synchronously (you won’t lose anything to a network drop) and then to Firestore in the background.

---

​

Bottom navigation tabs

Below the save row, the standard 5-tab navigation is still accessible:

The Capture Screen is a full-screen modal, so when you tap another tab the app stays in the same inspection — switching tabs doesn’t lose your in-progress finding.

---

​

What gets captured automatically

Several things happen behind the scenes during capture that you don’t have to manage:

---

​

Errors and recovery

---

​

Tips from inspectors

• Set section/subsection FIRST, then capture. Easier than relocating findings later.
• One photo = one finding for most deficiencies — keeps the report clean.
• Use Vision Scan only when the photo really tells the story. For a serial-number photo, plain Sys Info is faster and cheaper.
• Caption first, narrative last. AI Generate works best when the caption has shape. Don’t pre-fill the narrative — let AI populate it.
• Save + New aggressively. Each saved finding is durable and synced. Avoid stacking 30 in-memory findings in case the app gets backgrounded.
• For long inspections, use Shrink mode to give the form more room while still seeing the photo for reference.

---

​

Related articles

• Photos & Media — full media capture workflow
• AI During Capture — every AI tool in detail
• Photo Markup — annotating captured photos
• Ratings Screen — the fast-rate view for confirming all subsections are rated
• Review Screen — finding-by-finding editing of everything captured here
• Mobile Inspection Walkthrough — end-to-end practice