Protocol-level facts for Roboflow REST and Inference APIs — URL patterns, auth, parameters, error codes, and SDK quick-start. For deployment strategy and Workflow execution patterns, see roboflow-inference.
90
—
Does it follow best practices?
Impact
98%
1.42xAverage score across 3 eval scenarios
Risky
Do not use without reviewing
Python inference script with correct SDK usage and bounding box coordinate parsing
Uses inference-sdk package
0%
100%
InferenceHTTPClient usage
0%
100%
api_url constructor arg
0%
100%
api_key constructor arg
0%
100%
Correct serverless host
0%
100%
model_id format
0%
100%
Correct top-left x1
100%
100%
Correct top-left y1
100%
100%
Correct bottom-right x2
100%
100%
Correct bottom-right y2
100%
100%
requirements.txt lists inference-sdk
0%
100%
Scoped CI/CD API key setup and rotation script
Minimal scope only
100%
100%
Explicit scopes array
100%
100%
Immediate secret capture
100%
100%
Protected flag set
100%
100%
Env file storage
70%
90%
Never commit reminder
100%
75%
Disable before revoke
16%
100%
Zero-downtime rotation order
100%
100%
Correct CLI syntax
100%
100%
Management auth method
100%
100%
Roboflow image upload and training pipeline script
Correct upload SDK
100%
100%
Correct project creation fields
100%
100%
Valid project type
100%
100%
Correct upload host
100%
100%
Split parameter usage
100%
100%
Multiple splits used
100%
100%
Training status via version GET
100%
83%
API key authentication
100%
100%
Image size limit documented
0%
100%
Duplicate skip behavior documented
100%
100%
Training status polling explanation
100%
100%
bf3aefb
Table of Contents
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.