product-analytics

Pass

Audited by Gen Agent Trust Hub on Apr 7, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The SKILL.md file includes a 'Companion check' section that directs the agent to execute a shell command using ls on multiple paths such as ~/.claude/skills/ and ~/.agent/skills/ to discover which recommended companion skills are currently installed on the user's machine.
  • [EXTERNAL_DOWNLOADS]: Both the README.md and SKILL.md suggest that the user or agent install additional software from an external repository using the command 'npx skills add AbsolutelySkilled/AbsolutelySkilled'. While this originates from the identified vendor AbsolutelySkilled, it involves downloading and executing external packages.
  • [PROMPT_INJECTION]: The skill presents an indirect prompt injection vulnerability surface because its primary purpose is to ingest and process product data from potentially untrusted sources while maintaining shell access capabilities.
  • Ingestion points: The skill is designed to analyze user-provided event logs, cohort tables, and analytics reports which are processed in the agent context.
  • Boundary markers: No specific delimiters or instructions (e.g., 'ignore embedded commands') are provided to separate data from instructions during analysis.
  • Capability inventory: The agent has the capability to execute shell commands as demonstrated by the companion check instruction.
  • Sanitization: The instructions do not define any sanitization, validation, or escaping routines for the external data being analyzed.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 7, 2026, 03:05 PM
Security Audit — agent-trust-hub — product-analytics