yasarshaikh/SF-bench

benchmark-runner

Guides evaluation workflow execution for SF-Bench. The agent invokes this skill when running evaluations, discussing scoring methodology, or working with the evaluation pipeline.

Langue du texte source : anglais

code-reviewer

Performs code reviews following SF-Bench quality standards. The agent invokes this skill when reviewing code, evaluating pull requests, or assessing code quality for production-readiness.

Langue du texte source : anglais

salesforce-expert

Provides Salesforce platform expertise for Apex, LWC, Flow, and declarative development. The agent invokes this skill when working on Salesforce code, discussing platform constraints, or troubleshooting Salesforce-specific issues.

Langue du texte source : anglais

self-reviewer

Performs independent self-review of completed work to catch issues before they become problems. The agent invokes this skill when completing tasks, reviewing changes, or validating quality.

Langue du texte source : anglais

troubleshooter

Diagnoses and resolves evaluation failures, patch application errors, and Salesforce issues. The agent invokes this skill when debugging problems, investigating errors, or troubleshooting failures.

Langue du texte source : anglais