ohou-today-deal

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill fetches instructions and legal disclaimers using npx -y @nomadamas/k-skill@0. This represents a download of the vendor's own CLI tool from the npm registry to manage skill execution.
  • [REMOTE_CODE_EXECUTION]: The skill utilizes npx to dynamically download and execute the @nomadamas/k-skill package. This is used for both instructional purposes and to bridge the execution of the Python logic.
  • [COMMAND_EXECUTION]: The instructions direct the agent to run shell commands to interact with the Python script and the vendor's CLI tool to list deals or verify offline fixtures.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest data from an external retail website, which constitutes a surface for indirect prompt injection.
  • Ingestion points: The file scripts/ohou_today_deal.py fetches the public "Today's Deal" HTML page from Ohouse.
  • Boundary markers: The raw HTML is parsed to extract structured __NEXT_DATA__ JSON, which is then filtered and returned to the agent as a clean JSON object.
  • Capability inventory: The skill's capabilities are restricted to network GET requests and data parsing. It contains no functionality for file system modification, credential access, or arbitrary command execution based on the ingested content.
  • Sanitization: External content is processed strictly via Python's standard json.loads library, ensuring the data is not treated as executable instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 07:12 AM
Security Audit — agent-trust-hub — ohou-today-deal