Ingests drone imagery into SAR incidents by tiling frames and filtering them into manifests.
npx skills add https://github.com/sarocu/tile-search --skill sar-ingest
Please help me install this Agent Skill. Skill: sar-ingest Source: https://github.com/sarocu/tile-search/tree/main/.claude/skills/sar-ingest Command: npx skills add https://github.com/sarocu/tile-search --skill sar-ingest
Ingest drone imagery into a triage-ready SAR incident by enumerating frames, tiling each frame into a standardized grid, and pre-filtering tiles for rapid review.
Ingest the incident directory (e.g. assets/incident013) to produce runs/incident013/manifest.json.
High-intent search queries and answers about installing and using this skill.
To ingest drone imagery for search and rescue triage, enumerate high-resolution frames in a directory, extract EXIF GPS data, tile each frame into a 4x3 grid, and apply Stage-1 filters to produce a canonical manifest for downstream review.
Tiling drone imagery into a 4x3 grid breaks large frames into 1164x1147 tiles for rapid review, allowing a Stage-1 filter to mark keeps or skips and generating a structured manifest for triage-ready SAR incidents.
Yes, you can extract GPS coordinates and timestamps from drone images during the ingest process by reading EXIF data from high-resolution frames and writing them into a canonical runs manifest for reporting.
No, you do not need a specific incident profile to process drone image sets; the ingest process loads a canonical incident.json profile if available, or automatically uses a default profile when requested to generate the manifest.
The best way to prepare large drone image sets for SAR evaluation is to enumerate the frames, tile them into standardized grids, apply pre-filtering to mark keeps or skips, and output a canonical manifest with per-frame tile images.
Some drone image tiles are skipped during the SAR ingest process because a Stage-1 filter evaluates the 4x3 grid tiles and marks certain tiles as skips to reduce the dataset size for rapid triage review.
curiositechProcess geospatial data with PostGIS, GDAL, and npm packages.
sarocuReview SAR drone tiles in batches against incident-specific cues and log findings to findings.jsonl.
dkyazzentwatwaExtract and manage EXIF metadata from image files.
xdgExtract specific fields from large YAML files using yq syntax.
