Automatic Number Plate Recognition

Detect and read the largest license plate from an image using the TrafficEye REST API. Use when the user wants ANPR, ALPR, license plate OCR, number plate re...

Security Scan

The skill's requirements, instructions, and included code are consistent with a TrafficEye ANPR client; nothing in the bundle suggests hidden or unrelated behavior, though uploading images to an external API is a privacy consideration you should review.

Purpose & Capability

Name/description match the code and runtime instructions: the helper uploads a local image to the TrafficEye recognition API and parses plate results. Requested env vars (API key, API URL, auth mode, field names, request JSON, timeout) are reasonable for a configurable HTTP client.

Instruction Scope

SKILL.md instructs the agent to verify a local image path and call the provided Python helper which only reads the image (or a provided sample JSON) and posts multipart/form-data to the configured API URL. The instructions do not ask the agent to read unrelated files or secrets.

Install Mechanism

No install spec; this is essentially instruction + included Python script. No downloads, obscure installers, or archive extraction. The included source is readable (no obfuscation) and uses standard urllib APIs.

Credentials

The skill declares several configuration env vars (API key, API URL, auth mode/name, file/request field names, request JSON, timeout). This is somewhat verbose but matches a highly configurable client; the primary credential is just the API key, which is expected. Ensure you only provide a TrafficEye key you trust and do not re-use highly privileged credentials.

Persistence & Privilege

The skill does not request always:true, does not persist changes to other skills or system config, and allows autonomous invocation by default (normal for skills). There is no code that modifies agent configuration or other skills.

Assessment

This skill appears to be a legitimate TrafficEye ANPR client and includes its Python source for review. Before installing: (1) Confirm you trust trafficeye.ai and the API key you will supply — images uploaded may contain personal data (license plates, vehicle context); (2) Keep the API key limited in scope and not reused for other services; (3) Verify TRAFFICEYE_API_URL is the official endpoint (an attacker could redirect traffic if you set that variable to an untrusted host); (4) If you need offline processing or stronger privacy guarantees, avoid sending sensitive images to a cloud API; (5) Review the included recognize_plate.py if you want to be certain no additional data is collected or logged.

TrafficEye License Plate Reader

Use this skill when the user wants to read a license plate from an image with the TrafficEye API.

What This Skill Does

1. Accepts a local image path.
2. Uploads the image to the TrafficEye recognition API.
3. Optionally sends a request form field if TRAFFICEYE_REQUEST_JSON is configured.
4. Parses the API response.
5. Picks the largest detected plate by polygon area.
6. Returns the full selected plate payload to the user, including text, type (country), dimension, scores, occlusion, unreadable, and position.

Expected Input

• A local image file path.
• If the user supplied an attachment instead of a path, first resolve it to a local file path and then run the helper.

Default Runtime Assumptions

• The API endpoint defaults to https://trafficeye.ai/recognition.
• The default request payload is {"tasks":["DETECTION","OCR"],"requestedDetectionTypes":["BOX","PLATE"]}.
• The default API-key transport matches the TrafficEye public API example: header mode with header name apikey.
• Auth and request fields remain configurable in case your deployment differs.

Environment Variables

• TRAFFICEYE_API_KEY: required unless passed explicitly to the helper.
• TRAFFICEYE_API_URL: optional, defaults to https://trafficeye.ai/recognition.
• TRAFFICEYE_API_KEY_MODE: one of header, bearer, form, query. Default: header.
• TRAFFICEYE_API_KEY_NAME: key name for header, form, or query mode. Default: apikey.
• TRAFFICEYE_FILE_FIELD: multipart field for the image. Default: file.
• TRAFFICEYE_REQUEST_FIELD: multipart field for the JSON request. Default: request.
• TRAFFICEYE_REQUEST_JSON: JSON string to include as the request field. By default this is {"tasks":["DETECTION","OCR"],"requestedDetectionTypes":["BOX","PLATE"]}.
• TRAFFICEYE_TIMEOUT_S: optional timeout in seconds. Default: 30.

How To Run

Setup your API key:

export TRAFFICEYE_API_KEY='YOUR_REAL_KEY'

Use the bundled helper:

python3 recognize_plate.py /absolute/path/to/image.jpg

For structured output:

python3 recognize_plate.py /absolute/path/to/image.jpg --format json

If the deployment expects Bearer auth:

TRAFFICEYE_API_KEY_MODE=bearer python3 recognize_plate.py /absolute/path/to/image.jpg

If the deployment needs an explicit request payload:

TRAFFICEYE_REQUEST_JSON='{"requestedDetectionTypes":["PLATE"]}' python3 recognize_plate.py /absolute/path/to/image.jpg --format json

Equivalent to the documented public API example:

curl -X POST \
  -H "Content-Type: multipart/form-data" \
  -H "apikey: YOUR_API_KEY_HERE" \
  -F "file=@image.jpg" \
  -F 'request={"tasks":["DETECTION","OCR"],"requestedDetectionTypes":["BOX","PLATE"]}' \
  https://trafficeye.ai/recognition

Agent Workflow

1. Verify that the image path exists.
2. Run python3 recognize_plate.py <image-path> --format json.
3. Present the full selected plate payload to the user, especially text, type, dimension, occlusion, unreadable, and position.
4. If the API returns no readable text, explain that the largest plate was found but OCR text was missing.
5. If authentication fails, ask the user which auth mode their deployment expects and retry with the matching environment variables.

Offline Validation

You can validate the selection logic without calling the API:

python3 recognize_plate.py --response-json-file examples/sample_response.json --format json

Notes

• The helper intentionally chooses the largest plate by geometric area, not by detection confidence.
• The response parser first checks combinations[].roadUsers[].plates[], then also supports roadUsers[].plates[], top-level plates[], and nested plate payloads discovered recursively.
• The default request and auth header mirror the public example at https://www.trafficeye.ai/api.
• The selected result now includes the original plate payload from the API so country/type and all scores are preserved.