oracle-facet

Pass

Audited by Gen Agent Trust Hub on Sep 10, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill contains a shell script in the 'Step 0' section designed to discover profile libraries. The script uses bash -c to traverse multiple local directories, including paths within the user's home folder ($HOME/.claude/skills/, $HOME/.claude/personas/), and uses ls and sed to list available persona files.
  • [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to ingest and analyze external data (files, URLs, or documents) as 'artifacts' to be reviewed by a specific persona, which constitutes an attack surface for indirect prompt injection.
  • Ingestion points: Files, URLs, or decks provided by the user for analysis in 'Step 2'.
  • Boundary markers: The instructions do not specify any delimiters or safety markers to prevent the agent from following instructions embedded within the analyzed artifacts.
  • Capability inventory: The skill context supports shell execution (bash), file system access, and web search tools.
  • Sanitization: No sanitization or escaping of the ingested artifact content is performed before the agent processes and interprets the data.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 10, 2026, 05:24 AM
Security Audit — agent-trust-hub — oracle-facet