dotnet-ci-benchmarking
Pass
Audited by Gen Agent Trust Hub on Mar 5, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The workflow examples utilize official and trusted GitHub Actions from the
actionsorganization for repository checkout, .NET setup, and artifact management. - [PROMPT_INJECTION]: Identifies a potential surface for indirect prompt injection within the automated PR commenting process.
- Ingestion points: Benchmark results are read from JSON files (
*-report-full.json) in theload_benchmarksfunction of thecompare-benchmarks.pyscript. - Boundary markers: The workflow does not employ boundary markers or instructions to ignore embedded content when generating the PR comment body from the benchmark report.
- Capability inventory: The skill leverages the
actions/github-scriptaction to post content from thebenchmark-comparison.mdfile directly to Pull Request comments. - Sanitization: The provided Python script interpolates benchmark names and statistics into a markdown table without sanitization, which could lead to markdown or HTML injection if benchmark metadata is manipulated by an untrusted source.
Audit Metadata