Matlab Driving Data Importer
Wrap raw driving sensor logs (GPS, lidar, camera, tracks) into MATLAB scenariobuilder objects
Test report
- Verdict
- Works with setup
- Score
- Tested
- Jul 21, 2026
- Environment
- Claude Code 2.x (agent harness)
- Upstream re-checked
- Aug 10, 2026 · 8188742
Fetched SKILL.md + manifest.yaml and both referenced files (references/sensor-import-api.md, references/workflow-driving-log-analyzer.md) raw — both HTTP 200. Frontmatter parses with name+description; no hardcoded machine paths; security scan for curl|sh, base64, credential exfil, and injection strings found nothing (references even warn against fabricating nonexistent wrappers). Output test: task "load a driving-log CSV (time,lat,lon,alt) GPS into a scenariobuilder object for later scenario building." Baseline artifact: t=readtable(...); gpsData=scenariobuilder.GPSData(t.time,t.lat,t.lon,t.alt); and a suggestion to open drivingLogAnalyzer to verify. Skill artifact adds the canonical convertTimestamps(gpsData,"numeric") + timeRef=normalizeTimestamps(gpsData) pair, altitude-zeroing note, inspect-first step, and per Rule 0 does NOT auto-launch DLA — saves to sandbox and hands off to matlab-scenario-builder. Concrete difference is real, but I cannot run MATLAB/Automated Driving Toolbox/Scenario Builder to confirm the code executes or that the normalization actually prevents the claimed downstream failures, so outputMeasured=false and verdict=setup (blocking step: no MATLAB runtime).
Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works
- Installs cleanly 5/5
- Triggers reliably 5/5
- Output vs. baseline 6/10
- Docs & honesty 4/5
What Matlab Driving Data Importer does
Guides Claude to import recorded driving sensor data (GPS, camera, lidar, actor tracks, lanes) from datasets like KITTI, nuScenes, Waymo, ROS bags, or CSV into MATLAB scenariobuilder.* objects, then synchronize, crop, offset-correct, and normalize timestamps before scenario building. Triggers when a user has raw driving/vehicle/sensor logs to wrap, asks to sync or align multi-sensor recordings, or names drivingLogAnalyzer/DLA; it hands off scenario generation and RoadRunner export to matlab-scenario-builder. Requires MATLAB with Automated Driving Toolbox and Scenario Builder.
How to install Matlab Driving Data Importer
git clone --depth 1 https://github.com/matlab/matlab-agentic-toolkit.git /tmp/matlab-driving-data-importer-src
mkdir -p ~/.claude/skills
cp -R /tmp/matlab-driving-data-importer-src/skills-catalog/automotive/matlab-driving-data-importer ~/.claude/skills/matlab-driving-data-importer
# Requires MATLAB R2024b+ with Automated Driving Toolbox AND Scenario Builder for Automated Driving Toolbox.
# Meant to run inside the matlab-agentic-toolkit (needs evaluate_matlab_code / run_matlab_file / check_matlab_code tools).
# Ships two references (references/sensor-import-api.md, references/workflow-driving-log-analyzer.md) copied by the -R above.
Skills live in ~/.claude/skills/ (global) or .claude/skills/
(per-project). Restart Claude Code after installing.
Commands — how to trigger Matlab Driving Data Importer
-
/matlab-driving-data-importerWrap raw driving sensor logs (GPS, lidar, camera, tracks) into MATLAB scenariobuilder objects
It also activates on plain-language prompts like these:
-
Import this KITTI dataset into scenariobuilder objects -
Synchronize and crop this recorded lidar data -
Compute actor tracks from lidar with no annotations
Frequently asked questions
- Is the Matlab Driving Data Importer skill free?
- Yes. The skill itself is free from matlab/matlab-agentic-toolkit. SkillProof publishes the install command and an independent test verdict at no cost.
- Does Matlab Driving Data Importer work with Claude Code?
- We tested it with Claude Code 2.x (agent harness) on Jul 21, 2026. Verdict: Works with setup. Fetched SKILL.md + manifest.yaml and both referenced files (references/sensor-import-api.md, references/workflow-driving-log-analyzer.md) raw — both HTTP 200. Frontmatter parses with name+description; no hardcoded machine paths; security scan for curl|sh, base64, credential exfil, and injection strings found nothing (references even warn against fabricating nonexistent wrappers). Output test: task "load a driving-log CSV (time,lat,lon,alt) GPS into a scenariobuilder object for later scenario building." Baseline artifact: t=readtable(...); gpsData=scenariobuilder.GPSData(t.time,t.lat,t.lon,t.alt); and a suggestion to open drivingLogAnalyzer to verify. Skill artifact adds the canonical convertTimestamps(gpsData,"numeric") + timeRef=normalizeTimestamps(gpsData) pair, altitude-zeroing note, inspect-first step, and per Rule 0 does NOT auto-launch DLA — saves to sandbox and hands off to matlab-scenario-builder. Concrete difference is real, but I cannot run MATLAB/Automated Driving Toolbox/Scenario Builder to confirm the code executes or that the normalization actually prevents the claimed downstream failures, so outputMeasured=false and verdict=setup (blocking step: no MATLAB runtime).
- What is the Matlab Driving Data Importer SkillProof Score?
- 8.0/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 6/10, docs & honesty 4/5.
- How do I install Matlab Driving Data Importer?
- Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.
- Can I use Matlab Driving Data Importer with Cursor, Copilot, Gemini CLI, Codex or other AI tools?
- The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.