# AI use cases for Geo Sleuth (Oldcircle/geo-sleuth)

Repo: [Oldcircle/geo-sleuth](https://github.com/Oldcircle/geo-sleuth) | Use cases: 3 | Page: https://dedupedai.com/use-cases/oldcircle-geo-sleuth/

geo-sleuth is an agent skill made of Python scripts that estimates where a photo was taken, using OpenStreetMap geometry, elevation skylines, satellite imagery and street view, and reports its evidence.

This file holds every use case of this repo. Each one starts with a "## " heading and ends with its URL.

- Find where a photo without GPS was taken: https://dedupedai.com/use-cases/oldcircle-geo-sleuth/locate-photo-without-gps/
- Match a mountain skyline to a power line site: https://dedupedai.com/use-cases/oldcircle-geo-sleuth/match-photo-skyline-to-terrain/
- Pin a city street photo using public street-tree data: https://dedupedai.com/use-cases/oldcircle-geo-sleuth/pin-street-photo-with-street-tree-data/

## How to find where a photo without GPS was taken with Geo Sleuth

Repo: [Oldcircle/geo-sleuth](https://github.com/Oldcircle/geo-sleuth) | Difficulty: Intermediate | Time: 75 minutes | Basis: From the README

Give an agent a photo with no location data and get back camera position, heading and an evidence image.

**Who it is for:** An OSINT researcher or journalist holding a photo with stripped EXIF data who needs its location for a story or check.

**The problem:** A photo with no sign, plate or landmark gives no obvious clue, and manual search across maps and terrain takes hours.

**The outcome:** The agent returns the camera position with an error radius, the direction the camera faced, a satellite evidence image and a graded confidence.

### Steps

1. Run `npx skills add Oldcircle/geo-sleuth` and pick your agents when prompted.
2. Ask the agent to run doctor.py from the installed skill's scripts/ folder and fix any failed checks. (inferred, check the README)
3. Hand the agent the photo and say: find where this photo was taken.
4. Read the returned coordinates, error radius, heading and evidence image.
5. Open the evidence image and check that the named files and commands from the session support the conclusion.

### Prompt to paste into your agent

```text
Use the geo-sleuth skill from the Oldcircle/geo-sleuth repo to find where <path to photo> was taken. Expect a camera position with an error radius, a heading, a satellite evidence image and a confidence grade. Verify that the conclusion names the commands run and the files they produced.
```

### Caveat

The README's cases are single runs by one model, and the README says a different model or different choices may not reach the same result.

Risk, remote script: The install command fetches and runs the skill from GitHub, so review the repo before installing on a machine with sensitive files.

Risk, reverse engineering: The skill can pinpoint where a person was photographed, so use it only on photos you have the right to analyze and do not use it to track individuals.

Dual-use notice: only use this on software, sites and systems you own or are authorized to test.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/oldcircle-geo-sleuth/locate-photo-without-gps/

## How to match a mountain skyline to a power line site with Geo Sleuth

Repo: [Oldcircle/geo-sleuth](https://github.com/Oldcircle/geo-sleuth) | Difficulty: Advanced | Time: 45 minutes | Basis: From the README

Use nearby power lines and the ridge line in a photo to narrow a location in a region.

**Who it is for:** A researcher with a landscape photo that shows bare mountains and a few pylons but no text, roads or buildings.

**The problem:** Many sites share similar ridges, so a plain skyline search often leaves several candidates that score about the same.

**The outcome:** The agent narrows the candidates to a single site with a skyline score and a satellite check along the line of sight.

### Steps

1. Install the skill with `npx skills add Oldcircle/geo-sleuth` and confirm the checks in doctor.py pass.
2. Tell the agent which region the photo is in, such as a province or country, as a hint.
3. Have the agent run `osm.py geom` to pull the power lines for the region. (inferred, check the README)
4. Have the agent run `terrain.py scan` to find sites where mountains fill the view, then `terrain.py ridge` on the photo. (inferred, check the README)
5. Have the agent run `terrain.py fit` on the candidate sites and check the best site on satellite imagery. (inferred, check the README)

### Prompt to paste into your agent

```text
Use the geo-sleuth skill from the Oldcircle/geo-sleuth repo to narrow where <path to photo> was taken, using power lines and the ridge line in the photo. Expect a single candidate site with its fit score and a satellite image of the line of sight. Verify that the score and the files produced by terrain.py fit are cited in the answer.
```

### Caveat

The desert case needed a roll fit added to terrain.py fit for a 1 degree tilt, and the README says the default cluster setting can leave some hits unfit.

Risk, reverse engineering: Matching a photo's skyline to terrain can reveal the place where a person stood, so confirm the purpose before running it on private photos.

Dual-use notice: only use this on software, sites and systems you own or are authorized to test.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/oldcircle-geo-sleuth/match-photo-skyline-to-terrain/

## How to pin a city street photo using public street-tree data with Geo Sleuth

Repo: [Oldcircle/geo-sleuth](https://github.com/Oldcircle/geo-sleuth) | Difficulty: Advanced | Time: 60 minutes | Basis: From the README

Use a city's open street-tree records and street features to find the block in a street photo.

**Who it is for:** A researcher checking a street photo from a city that publishes street-tree inventories, such as Vancouver in the README case.

**The problem:** Cities have thousands of blocks, and a photo with common trees and furniture gives little to narrow the search by eye.

**The outcome:** The agent returns the likely block with a location error of a few metres, plus the trees and slope checks that supported it.

### Steps

1. Install the skill with `npx skills add Oldcircle/geo-sleuth`, then run doctor.py from the skill's scripts/ folder.
2. Give the agent the photo and name the city if you know it, such as Vancouver.
3. Ask the agent to filter the city's public street-tree data by species and trunk size, then group the results by block.
4. Ask the agent to check street view and slope on the shortlisted blocks.
5. Compare the chosen block against the photo's plates, streetlights and housing style.

### Prompt to paste into your agent

```text
Use the geo-sleuth skill from the Oldcircle/geo-sleuth repo to find the block where <path to street photo> was taken, starting from public street-tree data for <city>. Expect a block with an error of a few metres and a list of the trees and checks behind it. Verify that each claim names the file or command that supports it.
```

### Caveat

The street-tree case depends on that city publishing its tree inventory, and the README does not say which other cities provide the same data.

Risk, reverse engineering: Pinning a street photo to a block can identify where a person lives, so use it only for photos you are entitled to analyze.

Dual-use notice: only use this on software, sites and systems you own or are authorized to test.

Disclaimer: Third-party repos we don't own or vouch for. AI-generated use cases can be wrong: review everything and run at your own risk. Full text: https://dedupedai.com/disclaimer/

URL: https://dedupedai.com/use-cases/oldcircle-geo-sleuth/pin-street-photo-with-street-tree-data/