service-communication-audit

Pass

Audited by Gen Agent Trust Hub on Feb 27, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill's reference guide provides ripgrep (rg) command patterns to be executed by the agent to find communication boundaries within the user's source code.
  • [PROMPT_INJECTION]: The skill processes untrusted source code as part of its auditing function, making it vulnerable to indirect prompt injection if the code contains malicious instructions.
  • Ingestion points: Source code files are read and processed during the mapping and analysis steps defined in the skill and its guide.
  • Boundary markers: The audit request template uses basic Markdown headers but lacks robust delimiters or explicit instructions to the delegated agent to ignore embedded commands.
  • Capability inventory: The skill has access to local file searching via shell commands, model delegation through the teams tool, and ticket management via tk.
  • Sanitization: The skill does not implement sanitization or validation of the source code content before passing it to the delegated analysis agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 27, 2026, 10:53 AM