Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →CAN bus communication testing and validation for automotive networks Covers 9 topics across hil-sil domain. Includes 9 skill files covering .
.claude/skills/pangzhenying2025-automotive-hil-sil/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-17 | ✗→✓ | ▲ Improved | -17% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 158% | 0% |
| case-01 | ✗→✓ | ▲ Improved | 42% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 258% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 66% | 0% |
9 skill files covering hil-sil domain for automotive software engineering.
execution:
{{test_type}} --dbc {{dbc_file}} description: Execute CAN bus tests setup:
description: Configure CAN interface validation:
description: Validate CAN communication
cleanup:
description: Destroy actors execution:
description: Start simulation
{{carla_port}} conditions:
description: Run scenario
/carla/ego_vehicle/gnss /carla/ego_vehicle/imu description: Record sensor data
description: Monitor vehicle telemetry setup:
description: Connect to CARLA server
{{weather}} description: Load town and set weather
description: Spawn ego vehicle
description: Attach sensors to ego vehicle
description: Configure traffic manager
description: Initialize ROS 2 bridge validation:
description: Analyze collision events
description: Evaluate sensor coverage
description: Generate simulation report
execution:
description: Start baseline recording
description: Inject fault at specified point
conditions:
description: Monitor system response
description: Record fault response setup:
description: Initialize fault injection framework
--parameters {{fault_parameters}} description: Configure fault scenario
description: Load test oracle validation:
{{test_oracle}} description: Compare against test oracle
description: Analyze safety mechanisms
description: Generate FMEA report
execution:
2.0}}''' description: Control vehicle via ROS 2
description: Monitor sensor topics
description: Record simulation data setup:
description: Source ROS 2 environment
description: Launch Gazebo with world file
description: Spawn robot model
conditions:
description: Start ROS 2 bridge
description: Configure physics engine validation:
description: Check TF tree
description: Visualize in RViz
setup:
description: Configure HIL platform connection
description: Load ECU software to HIL platform
description: Initialize CAN/FlexRay/LIN interfaces
description: Configure I/O channels and signal routing
description: Load test scenario and stimulation profiles validation:
description: Verify HIL platform readiness
description: Run loopback tests on all interfaces
description: Check signal integrity and timing
execution:
description: Start sensor data stream
conditions:
description: Apply noise model
conditions:
description: Publish to ROS 2 setup_camera:
description: Configure camera sensor
--fps {{sensor_config.fps}} --fov {{sensor_config.fov}} description: Set image parameters setup_lidar:
description: Configure LiDAR sensor
--range {{sensor_config.range}} --points-per-second {{sensor_config.points_per_second}} description: Initialize point cloud generation setup_radar:
description: Configure radar sensor parameters
--fov {{sensor_config.fov}} --resolution {{sensor_config.resolution}} description: Set radar detection parameters setup_ultrasonic:
{{sensor_config}} description: Configure ultrasonic array
--beam-width {{sensor_config.beam_width}} description: Set ultrasonic detection cone validation:
description: Visualize sensor output
description: Validate sensor accuracy
analysis:
conditions:
description: Generate code coverage report
description: Analyze execution traces
description: Generate test report cleanup:
description: Stop virtual ECU
description: Remove virtual network interfaces execution:
description: Load test scenarios
description: Execute functional tests
description: Run regression tests
conditions:
description: Perform fault injection tests setup:
description: Create virtual ECU environment
description: Configure virtual CAN network
description: Start virtual ECU simulation
description: Initialize network stack and routing
{'description': 'Execute automated test suite', 'commands': 'python3 tests/automation/test_runner.py --suite {{test_suite}} --platform {{platform}}']}]
cleanup:
description: Stop simulation
description: Archive sensor data execution:
{{real_time_factor}} description: Start vehicle simulation
conditions:
description: Run sensor data collection
description: Execute scenario-based tests
description: Monitor ECU responses setup:
description: Initialize vehicle simulator
description: Configure sensor suite
description: Load driving scenario
description: Setup ECU integration bridges
description: Initialize ROS 2 communication stack validation:
description: Analyze sensor fusion accuracy
description: Validate vehicle dynamics behavior
description: Check ECU decision quality
description: Generate VIL test report
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-17 | fail→pass | 22,827 | 3,539 | -84% | 1 | 1 | 0% | 5,331 | 4,423 | -17% | 0 | 0 | — |
case-02 | fail→pass | 10,096 | 10,875 | +8% | 1 | 1 | 0% | 1,901 | 4,906 | +158% | 0 | 0 | — |
case-01 | fail→pass | 16,667 | 5,748 | -66% | 1 | 1 | 0% | 3,721 | 5,286 | +42% | 0 | 0 | — |
case-03 | fail→pass | 25,313 | 5,058 | -80% | 1 | 1 | 0% | 1,450 | 5,198 | +258% | 0 | 0 | — |
case-04 | pass→pass | 11,899 | 10,222 | -14% | 1 | 1 | 0% | 2,632 | 6,059 | +130% | 0 | 0 | — |
case-05 | pass→pass | 15,288 | 9,202 | -40% | 1 | 1 | 0% | 3,432 | 5,760 | +68% | 0 | 0 | — |
case-06 | pass→pass | 11,715 | 10,399 | -11% | 1 | 1 | 0% | 2,552 | 5,845 | +129% | 0 | 0 | — |
case-07 | fail→pass | 27,617 | 3,227 | -88% | 1 | 1 | 0% | 2,839 | 4,723 | +66% | 0 | 0 | — |
case-08 | fail→pass | 8,586 | 3,554 | -59% | 1 | 1 | 0% | 2,003 | 4,652 | +132% | 0 | 0 | — |
case-09 | fail→pass | 13,540 | 4,321 | -68% | 1 | 1 | 0% | 3,050 | 4,973 | +63% | 0 | 0 | — |
case-10 | fail→pass | 9,450 | 3,419 | -64% | 1 | 1 | 0% | 2,324 | 4,649 | +100% | 0 | 0 | — |
case-11 | fail→pass | 17,088 | 4,030 | -76% | 1 | 1 | 0% | 4,202 | 4,749 | +13% | 0 | 0 | — |
case-12 | fail→pass | 13,334 | 3,732 | -72% | 1 | 1 | 0% | 3,116 | 4,784 | +54% | 0 | 0 | — |
case-13 | fail→pass | 18,993 | 3,556 | -81% | 1 | 1 | 0% | 4,348 | 4,564 | +5% | 0 | 0 | — |
case-14 | fail→pass | 18,812 | 3,772 | -80% | 1 | 1 | 0% | 3,829 | 4,601 | +20% | 0 | 0 | — |
case-15 | fail→pass | 26,501 | 5,843 | -78% | 1 | 1 | 0% | 4,992 | 5,442 | +9% | 0 | 0 | — |
case-16 | fail→pass | 4,927 | 1,631 | -67% | 1 | 1 | 0% | 973 | 4,135 | +325% | 0 | 0 | — |
case-18 | fail→pass | 10,705 | 1,299 | -88% | 1 | 1 | 0% | 2,129 | 4,025 | +89% | 0 | 0 | — |
case-19 | pass→pass | 4,988 | 2,195 | -56% | 1 | 1 | 0% | 1,104 | 4,269 | +287% | 0 | 0 | — |
case-20 | fail→pass | 21,942 | 3,936 | -82% | 1 | 1 | 0% | 4,318 | 4,669 | +8% | 0 | 0 | — |
case-21 | fail→pass | 23,050 | 3,816 | -83% | 1 | 1 | 0% | 5,014 | 4,490 | -10% | 0 | 0 | — |
case-22 | fail→pass | 15,766 | 2,271 | -86% | 1 | 1 | 0% | 2,515 | 4,252 | +69% | 0 | 0 | — |
case-23 | fail→pass | 7,008 | 1,850 | -74% | 1 | 1 | 0% | 1,418 | 4,098 | +189% | 0 | 0 | — |
case-24 | fail→pass | 19,639 | 6,017 | -69% | 1 | 1 | 0% | 3,283 | 5,233 | +59% | 0 | 0 | — |
case-25 | pass→pass | 6,314 | 2,501 | -60% | 1 | 1 | 0% | 1,205 | 4,396 | +265% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 25 cases were attempted, and 24 counted toward the lift figure. The other 1 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +80 percentage points is the difference between those two pass rates over the 24 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.