dotnet-ci-benchmarking
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's workflow ingests benchmark metadata from machine-generated JSON reports and incorporates it into a Markdown summary that is automatically posted as a GitHub Pull Request comment. This creates a vulnerability where malicious code contributors could use crafted benchmark names to inject content into the GitHub UI.
- Ingestion points: Benchmark result JSON files located in
benchmarks/BenchmarkDotNet.Artifacts/results/are consumed by a Python script. - Boundary markers: There are no explicit boundary markers or instructions telling the processing agent to treat ingested benchmark names as untrusted data.
- Capability inventory: The skill utilizes
actions/github-scriptwith theactions: readandissues: write(via metadata) permissions to post comments directly to Pull Requests. - Sanitization: The
compare-benchmarks.pyscript provided in the skill performs no escaping or sanitization of theFullNamefield before placing it in the Markdown table structure.
Audit Metadata