yolo-detection-2026

Name: yolo-detection-2026
Author: SharpAI

2.9kstars460forks

How to install yolo-detection-2026?

sharpai/deepcamera/yolo-detection-2026

$npx -y skills add SharpAI/DeepCamera --skill yolo-detection-2026

Installs into the current project.

Summary

YOLO 2026 — state-of-the-art real-time object detection

Files · 1

View on GitHub

SKILL.md

1	# YOLO 2026 Object Detection
2
3	Real-time object detection using the latest YOLO 2026 models. Detects 80+ COCO object classes including people, vehicles, animals, and everyday objects. Outputs bounding boxes with labels and confidence scores.
4
5	## Model Sizes
6
7	\| Size \| Speed \| Accuracy \| Best For \|
8	\|------\|-------\|----------\|----------\|
9	\| nano \| Fastest \| Good \| Real-time on CPU, edge devices \|
10	\| small \| Fast \| Better \| Balanced speed/accuracy \|
11	\| medium \| Moderate \| High \| Accuracy-focused deployments \|
12	\| large \| Slower \| Highest \| Maximum detection quality \|
13
14	## Hardware Acceleration
15
16	The skill uses [`env_config.py`](../../lib/env_config.py) to automatically detect hardware and convert the model to the fastest format for your platform. Conversion happens once during deployment and is cached.
17
18	\| Platform \| Backend \| Optimized Format \| Compute Units \| Expected Speedup \|
19	\|----------\|---------\|------------------\|:-------------:\|:----------------:\|
20	\| NVIDIA GPU \| CUDA \| TensorRT `.engine` \| GPU \| ~3-5x \|
21	\| Apple Silicon (M1+) \| MPS \| CoreML `.mlpackage` \| Neural Engine (NPU) \| ~2x \|
22	\| Intel CPU/GPU/NPU \| OpenVINO \| OpenVINO IR `.xml` \| CPU/GPU/NPU \| ~2-3x \|
23	\| AMD GPU \| ROCm \| ONNX Runtime \| GPU \| ~1.5-2x \|
24	\| CPU (any) \| CPU \| ONNX Runtime \| CPU \| ~1.5x \|
25
26	> Apple Silicon Note: Detection defaults to `cpu_and_ne` (CPU + Neural Engine), keeping the GPU free for LLM/VLM inference. Set `compute_units: all` to include GPU if not running local LLM.
27
28	### How It Works
29
30	1. `deploy.sh` detects your hardware via `env_config.HardwareEnv.detect()`
31	2. Installs the matching `requirements_{backend}.txt` (e.g. CUDA → includes `tensorrt`)
32	3. Pre-converts the default model to the optimal format
33	4. At runtime, `detect.py` loads the cached optimized model automatically
34	5. Falls back to PyTorch if optimization fails
35
36	Set `use_optimized: false` to disable auto-conversion and use raw PyTorch.
37
38	## Auto Start
39
40	Set `auto_start: true` in the skill config to start detection automatically when Aegis launches. The skill will begin processing frames from the selected camera immediately.
41
42	```yaml
43	auto_start: true
44	model_size: nano
45	fps: 5
46	```
47
48	## Performance Monitoring
49
50	The skill emits `perf_stats` events every 50 frames with aggregate timing:
51
52	```jsonl
53	{"event": "perf_stats", "total_frames": 50, "timings_ms": {
54	"inference": {"avg": 3.4, "p50": 3.2, "p95": 5.1},
55	"postprocess": {"avg": 0.15, "p50": 0.12, "p95": 0.31},
56	"total": {"avg": 3.6, "p50": 3.4, "p95": 5.5}
57	}}
58	```
59
60	## Protocol
61
62	Communicates via JSON lines over stdin/stdout.
63
64	### Aegis → Skill (stdin)
65	```jsonl
66	{"event": "frame", "frame_id": 42, "camera_id": "front_door", "timestamp": "...", "frame_path": "/tmp/aegis_detection/frame_front_door.jpg", "width": 1920, "height": 1080}
67	```
68
69	### Skill → Aegis (stdout)
70	```jsonl
71	{"event": "ready", "model": "yolo2026n", "device": "mps", "backend": "mps", "format": "coreml", "gpu": "A

People who liked this also viewed

6 picks

Type

find-skillsHelps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skSkillsJul 20262.6M27k
grill-meA relentless interview to sharpen a plan or design.SkillsJul 2026633k182k
vercel-react-best-practicesVercel's official collection of agent skillsSkillsJul 2026573k29k
improve-codebase-architectureScan a codebase for deepening opportunities, present them as a visual HTML report, then grill through whichever one you pick.SkillsJul 2026522k182k
setup-matt-pocock-skillsConfigure this repo for the engineering skills — set up its issue tracker, triage label vocabulary, and domain doc layout. Run once before first use of the other engineering skills.SkillsJul 2026454k182k
triageMove issues and external PRs through a state machine of triage roles — categorise, verify, grill if needed, and write agent-ready briefs.SkillsJul 2026424k182k

yolo-detection-2026

2.9kstars460forks

View on GitHub

How to install yolo-detection-2026?

sharpai/deepcamera/yolo-detection-2026

$npx -y skills add SharpAI/DeepCamera --skill yolo-detection-2026

Installs into the current project.

Summary

YOLO 2026 — state-of-the-art real-time object detection

Files · 1

View on GitHub

SKILL.md

1	# YOLO 2026 Object Detection
2
3	Real-time object detection using the latest YOLO 2026 models. Detects 80+ COCO object classes including people, vehicles, animals, and everyday objects. Outputs bounding boxes with labels and confidence scores.
4
5	## Model Sizes
6
7	\| Size \| Speed \| Accuracy \| Best For \|
8	\|------\|-------\|----------\|----------\|
9	\| nano \| Fastest \| Good \| Real-time on CPU, edge devices \|
10	\| small \| Fast \| Better \| Balanced speed/accuracy \|
11	\| medium \| Moderate \| High \| Accuracy-focused deployments \|
12	\| large \| Slower \| Highest \| Maximum detection quality \|
13
14	## Hardware Acceleration
15
16	The skill uses [`env_config.py`](../../lib/env_config.py) to automatically detect hardware and convert the model to the fastest format for your platform. Conversion happens once during deployment and is cached.
17
18	\| Platform \| Backend \| Optimized Format \| Compute Units \| Expected Speedup \|
19	\|----------\|---------\|------------------\|:-------------:\|:----------------:\|
20	\| NVIDIA GPU \| CUDA \| TensorRT `.engine` \| GPU \| ~3-5x \|
21	\| Apple Silicon (M1+) \| MPS \| CoreML `.mlpackage` \| Neural Engine (NPU) \| ~2x \|
22	\| Intel CPU/GPU/NPU \| OpenVINO \| OpenVINO IR `.xml` \| CPU/GPU/NPU \| ~2-3x \|
23	\| AMD GPU \| ROCm \| ONNX Runtime \| GPU \| ~1.5-2x \|
24	\| CPU (any) \| CPU \| ONNX Runtime \| CPU \| ~1.5x \|
25
26	> Apple Silicon Note: Detection defaults to `cpu_and_ne` (CPU + Neural Engine), keeping the GPU free for LLM/VLM inference. Set `compute_units: all` to include GPU if not running local LLM.
27
28	### How It Works
29
30	1. `deploy.sh` detects your hardware via `env_config.HardwareEnv.detect()`
31	2. Installs the matching `requirements_{backend}.txt` (e.g. CUDA → includes `tensorrt`)
32	3. Pre-converts the default model to the optimal format
33	4. At runtime, `detect.py` loads the cached optimized model automatically
34	5. Falls back to PyTorch if optimization fails
35
36	Set `use_optimized: false` to disable auto-conversion and use raw PyTorch.
37
38	## Auto Start
39
40	Set `auto_start: true` in the skill config to start detection automatically when Aegis launches. The skill will begin processing frames from the selected camera immediately.
41
42	```yaml
43	auto_start: true
44	model_size: nano
45	fps: 5
46	```
47
48	## Performance Monitoring
49
50	The skill emits `perf_stats` events every 50 frames with aggregate timing:
51
52	```jsonl
53	{"event": "perf_stats", "total_frames": 50, "timings_ms": {
54	"inference": {"avg": 3.4, "p50": 3.2, "p95": 5.1},
55	"postprocess": {"avg": 0.15, "p50": 0.12, "p95": 0.31},
56	"total": {"avg": 3.6, "p50": 3.4, "p95": 5.5}
57	}}
58	```
59
60	## Protocol
61
62	Communicates via JSON lines over stdin/stdout.
63
64	### Aegis → Skill (stdin)
65	```jsonl
66	{"event": "frame", "frame_id": 42, "camera_id": "front_door", "timestamp": "...", "frame_path": "/tmp/aegis_detection/frame_front_door.jpg", "width": 1920, "height": 1080}
67	```
68
69	### Skill → Aegis (stdout)
70	```jsonl
71	{"event": "ready", "model": "yolo2026n", "device": "mps", "backend": "mps", "format": "coreml", "gpu": "A

People who liked this also viewed

6 picks

Type