skills/baarai/artex/rea/Gen Agent Trust Hub

rea

Pass

Audited by Gen Agent Trust Hub on Nov 21, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted binary files (e.g., ELF, PE, APK, Mach-O, WASM) and JavaScript bundles. This introduces an attack surface where an adversary could embed malicious instructions within the strings or metadata of a binary to influence the agent's behavior during analysis.
  • Ingestion points: Untrusted files are loaded from the host filesystem via the open_binary tool (file: SKILL.md).
  • Capability inventory: The skill possesses capabilities to extract strings, search for procedures, and map application graphs. It is specifically instructed to look for sensitive information like API keys and secrets (file: SKILL.md).
  • Boundary markers: The tool returns results as structured 'Evidence objects' which provides some context separation, but the content of the analysis is untrusted and directly influences the agent's subsequent facts and findings (file: SKILL.md).
  • Sanitization: There is no evidence of sanitization or filtering for the data extracted from the binaries before it is processed by the agent's language model.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 21, 2026, 10:04 AM