skills/jmagly/aiwg/regression-report/Gen Agent Trust Hub

regression-report

Pass

Audited by Gen Agent Trust Hub on Jun 22, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface due to its core function of processing data from external, potentially untrusted sources such as issue trackers and CI logs.
  • Ingestion points: Data is imported from bisect reports, baseline comparisons, GitHub issues, CI/CD logs, and Datadog monitoring metrics.
  • Boundary markers: The skill does not explicitly define delimiters (e.g., XML tags) or instructions to ignore embedded commands when interpolating external data into its reports.
  • Capability inventory: The skill's capabilities are limited to reading local files and writing Markdown reports to the .aiwg/ directory. It has no tools for network exfiltration, shell command execution, or system modification.
  • Sanitization: No specific data validation or sanitization routines are mentioned for the ingested content.
  • Assessment: This surface is an inherent part of the skill's intended reporting functionality and is considered low-risk given the absence of any dangerous tools or exfiltration paths.
  • [SAFE]: No evidence of hardcoded credentials, code obfuscation, or persistence mechanisms was found. All external references point to legitimate developer services (GitHub, Datadog) or internal project paths, and the behavior is consistent with the skill's stated purpose.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 22, 2026, 06:30 PM
Security Audit — agent-trust-hub — regression-report