kamis-food-price

Pass

Audited by Gen Agent Trust Hub on Oct 22, 2026

Risk Level: SAFEREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill instructs the user to run npx -y @nomadamas/k-skill@0, which downloads and executes a remote package from the npm registry. This is used for managing skill instructions and updates, and the package is authored by the skill vendor.
  • [EXTERNAL_DOWNLOADS]: Fetches executable code from the npm registry and queries price data from a vendor-managed proxy domain.
  • [COMMAND_EXECUTION]: Provides several shell command examples (npx, curl) for the agent to execute in order to initialize the skill and retrieve food price data.
  • [DATA_EXFILTRATION]: The Python script run_kamis.py accesses a local configuration file at ~/.config/k-skill/secrets.env to load API keys. While this is used for legitimate API calls to KAMIS, reading files in the user's configuration directory is a data access pattern to be noted.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes food price data from the KAMIS API and a proxy service, which are external data sources.
  • Ingestion points: Data is fetched via urllib.request in run_kamis.py and curl in instruction.md.
  • Boundary markers: No explicit instruction delimiters are used to wrap the incoming price data.
  • Capability inventory: The skill can perform network requests and read local configuration files.
  • Sanitization: The script uses json.loads and specific key extraction to structure the data before outputting it.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 22, 2026, 09:48 PM