pp-ticketdata

Pass

Audited by Gen Agent Trust Hub on Jul 10, 2026

Risk Level: SAFEDATA_EXFILTRATIONEXTERNAL_DOWNLOADSPROMPT_INJECTIONREMOTE_CODE_EXECUTIONCOMMAND_EXECUTION
Full Analysis
  • [DATA_EXFILTRATION]: The CLI includes an output delivery feature (--deliver webhook:<url>) that enables transmitting command results to arbitrary external endpoints.
  • [EXTERNAL_DOWNLOADS]: The skill instructs the installation of binaries and packages from the author's official GitHub and NPM repositories (@mvanhorn).
  • [PROMPT_INJECTION]: The 'Learning Loop' and 'Playbook' features represent an indirect prompt injection surface. 1. Ingestion points: External event names, performer slugs, and price signals from the TicketData API, plus user strings in recall and teach commands. 2. Boundary markers: No specific delimiters or 'ignore' instructions are described for isolating untrusted API data. 3. Capability inventory: Shell command execution, local file/database writes, and network POST requests via webhooks. 4. Sanitization: No specific sanitization or validation steps are mentioned for data incorporated into stored playbooks or flag aliases.
  • [REMOTE_CODE_EXECUTION]: The 'Playbook' mechanism involves dynamic assembly and execution of shell commands using slot substitution ({slot}) based on stored patterns.
  • [COMMAND_EXECUTION]: The skill functions by executing the ticketdata-pp-cli and ticketdata-pp-mcp binaries for all documented tracking and analysis tasks.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 10, 2026, 01:30 AM
Security Audit — agent-trust-hub — pp-ticketdata