search

AI Based Binary Analysis in Your CI/CD

hashtag
High-Level Integration

The integration works by inserting two key steps into your existing build workflow:

1

hashtag
The upload Step

This step occurs immediately after your build process.

  • Action: Takes your compiled binaries (executables, libraries, etc.) and uploads them to the secure LOCI backend for analysis. Link: https://docs.loci-dev.net/cicd/#1-the-upload-step

  • Result: Analysis begins immediately in the background. If the LOCI GitHub App is installed on your repo, this step is sufficient to trigger automatic PR comments with results upon analysis completion.

2

hashtag
The summary Step

This optional step is used to pull the analysis results back into the workflow UI.

  • Action: Pauses the job briefly, waits for the LOCI analysis to finish, and retrieves the results. Link: https://docs.loci-dev.net/cicd/#2-the-summary-step

  • Result: Embeds the Agent Report and Function Insights directly into your GitHub workflow run summary. This makes critical performance data visible right where your developers review build status.

hashtag
Key Benefits in CI/CD

Feature
Description
Benefit

Pre-Test Analysis

Predicts issues from static analysis of the binary.

Catches hardware-software interaction problems before resource-intensive testing.

Autonomous Optimization

LOCI's AI provides concrete, actionable suggestions.

Reduces engineering effort required to diagnose and resolve performance bottlenecks.

Performance Gate

(Via GitHub App) Adds a Performance Review Check Run.

Automatically enforces performance policies, allowing the PR to fail if performance regressions are detected.

By integrating LOCI, your CI/CD pipeline gains an intelligent, autonomous layer of quality assurance focused specifically on CPU performance, power, and hardware reliability.

PreviousInter-Binary Relations in the Call GraphNextGitHub Integration

Last updated