Skills
Navigate
skills/posthog/skills/creating-ai-subscription/Gen Agent Trust Hub

creating-ai-subscription

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill handles untrusted free-text input via the prompt parameter, which is processed by an LLM to generate HogQL queries and reports. This presents a surface for indirect prompt injection.\n
  • Ingestion points: The prompt argument in the posthog:subscriptions-create tool in SKILL.md.\n
  • Boundary markers: No specific delimiters or safety instructions are defined in the skill to prevent the LLM from potentially following malicious instructions embedded within the prompt text.\n
  • Capability inventory: The skill uses posthog:subscriptions-create to schedule tasks that run LLM-generated HogQL queries over project data and deliver synthesised markdown reports to email or Slack.\n
  • Sanitization: No evidence of sanitization or validation for the prompt content is described before it is passed to the LLM-driven query generation pipeline.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 01:03 AM
Security Audit — agent-trust-hub — creating-ai-subscription