skills/reddb-io/red-skills/triage/Gen Agent Trust Hub

triage

Pass

Audited by Gen Agent Trust Hub on May 16, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests untrusted data from issue reports and comments to guide its recommendations.
  • Ingestion points: Reads issue bodies, comments, labels, and reporter history from the project's issue tracker (SKILL.md).
  • Boundary markers: Absent; the instructions do not define delimiters or provide specific 'ignore embedded instructions' warnings for the data being parsed.
  • Capability inventory: The skill can post comments, apply labels, close issues, and write markdown files to the .out-of-scope/ directory (SKILL.md, OUT-OF-SCOPE.md).
  • Sanitization: Absent; no explicit sanitization or validation of external issue content is performed before interpolation into the triage logic.
  • [COMMAND_EXECUTION]: The skill references and uses internal platform commands such as /setup-red-skills for label mapping and /start for interactive sessions. These appear to be vendor-specific tools provided by 'reddb-io'.
Audit Metadata
Risk Level
SAFE
Analyzed
May 16, 2026, 05:33 PM
Security Audit — agent-trust-hub — triage