CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

yolo-detection-2026-coral-tpu-macos

Google Coral Edge TPU — real-time object detection natively (macOS / Linux)

Quality

69%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Critical

Do not install without reviewing

Fix and improve this skill with Tessl

tessl review fix ./skills/detection/yolo-detection-2026-coral-tpu-macos/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

71%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is a concise, well-structured overview with concrete deploy commands and a precise JSONL protocol, supported by a clear inference workflow and a deploy-time fallback. Its main weakness is progressive disclosure: the repeatedly referenced deploy.sh is missing from the bundle and references are not signaled with links.

Suggestions

Add the missing deploy.sh to the scripts/ bundle (or correct the reference to the actual installer script such as install_pycoral.py) so referenced paths resolve.

Signal file references with markdown links (e.g., [detect.py](scripts/detect.py), [deploy script](scripts/deploy.sh)) for clearer navigation.

Tighten the Important Deployment Notice wording to remove redundant phrasing ('securely ... securely').

DimensionReasoningScore

Conciseness

The body is largely lean with a focused ASCII diagram, a performance table, and JSONL protocol examples; minor instances of over-explanation remain (e.g., the wordy deployment notice 'will natively halt execution and prompt you securely for your OS sudo password to securely register'). It is below 5 because those padded phrases could be trimmed, and above 3 because it mostly assumes Claude's competence without explaining basics.

4 / 5

Actionability

Provides concrete executable commands ('./deploy.sh') and a complete JSONL protocol with example payloads, but the actual detection invocation via the entry script is not shown inline (it lives in scripts/detect.py); below 5 because copy-paste usage of detect.py is not covered, above 3 because what is shown is real and executable rather than pseudocode.

4 / 5

Workflow Clarity

A clear 4-step sequence (frame write → stdin event → detect.py inference → stdout detections) is present, and the deploy step includes a fallback checkpoint when the sudo prompt is refused; minor validation gaps remain (no inline TPU-availability verification before inference). Not 5 because validation checkpoints are not fully explicit, not 3 because a real fallback/error path exists.

4 / 5

Progressive Disclosure

The body is a well-sectioned overview pointing to bundle scripts (detect.py is referenced and exists), but './deploy.sh' is referenced repeatedly and is absent from the scripts/ bundle, and file references are mentioned in prose rather than signaled with markdown links; below 4 because of the broken reference and lack of clear link signaling, above 2 because sections are organized and references are one level deep.

3 / 5

Total

15

/

20

Passed

Description

53%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description clearly conveys a specific hardware-accelerated capability but omits any explicit 'Use when...' trigger guidance, capping completeness and trigger quality. It is distinctive within its niche though it overlaps slightly with a sibling detection skill.

Suggestions

Add an explicit 'Use when...' clause naming trigger phrases (e.g., 'Use when the user mentions Coral/Edge TPU, real-time object detection, or on-device camera detection').

Include a few natural synonyms users might say (e.g., 'YOLO', 'computer vision', 'bounding-box detection') to broaden trigger coverage.

Distinguish from the sibling skill by noting the on-device/native hardware angle directly in the description.

DimensionReasoningScore

Specificity

Names the domain ('Google Coral Edge TPU') and one concrete action ('real-time object detection natively'), matching the anchor for domain plus 1-2 concrete actions without comprehensive coverage; it is above a score of 2 because the action is specific rather than generic, but below 4 because only one action is listed.

3 / 5

Completeness

Clearly answers 'what' (real-time object detection on Coral Edge TPU) but provides no 'when'/'Use when...' trigger guidance, so per the cap it cannot exceed 3; it is above 2 because the 'what' is explicit rather than vague.

3 / 5

Trigger Term Quality

Includes relevant natural terms like 'real-time object detection' and the hardware name 'Coral Edge TPU' that users would say, but misses common synonyms/variations such as 'YOLO', 'computer vision', 'bounding boxes', or 'object tracker'; not a 4 because coverage is incomplete.

3 / 5

Distinctiveness Conflict Risk

The Coral Edge TPU hardware-specific niche is mostly distinct with minimal conflict risk, but there is minor overlap risk with the closely related general 'yolo-detection-2026' skill referenced in the body; not a 5 because that sibling-skill overlap exists.

4 / 5

Total

13

/

20

Passed

Validation

93%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

15

/

16

Passed

Repository

SharpAI/DeepCamera

Path

skills/detection/yolo-detection-2026-coral-tpu-macos/SKILL.md

Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.