pp-airbnb

Pass

Audited by Gen Agent Trust Hub on Sep 18, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the agent to install the airbnb-pp-cli and airbnb-pp-mcp tools using npx and go install. These resources are fetched from the author's repository (github.com/mvanhorn/printing-press-library) and NPM scope (@mvanhorn/printing-press-library), which is consistent with the skill's authorship.
  • [DATA_EXFILTRATION]: The skill documentation describes a --deliver webhook:<url> flag and a feedback command. These features allow the agent to send command outputs or feedback notes to external web addresses. While these are legitimate tool features, they represent a capability to move data out of the local environment.
  • [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to scrape and process data from external websites (Airbnb and VRBO). Since this data is controlled by third parties, there is an inherent risk that a malicious listing could contain instructions intended to manipulate the AI agent.
  • Ingestion points: Data is ingested through commands such as airbnb_listing get and vrbo_listing search in SKILL.md.
  • Boundary markers: The skill does not explicitly define delimiters for scraped content in its instructions.
  • Capability inventory: The agent has access to the Bash tool and can perform network operations via the airbnb-pp-cli.
  • Sanitization: There is no evidence of specific sanitization or filtering logic applied to the scraped web content before it is returned to the agent.
  • [COMMAND_EXECUTION]: The skill relies on executing the airbnb-pp-cli binary via the Bash tool to perform its tasks. The instructions provide the specific command patterns the agent should use.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 18, 2026, 06:33 PM