implementing-api-surveys
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides documentation and implementation guidance for PostHog surveys. It correctly distinguishes between public project tokens and personal API keys.
- [INDIRECT_PROMPT_INJECTION]: The skill processes survey definitions from the PostHog API to generate implementation code. This represents a data ingestion surface where external content influences agent-generated code. 1. Ingestion points: Survey definitions fetched from the PostHog API via posthog:survey-get. 2. Boundary markers: No explicit delimiter or instruction to ignore embedded content is mentioned for the survey data itself. 3. Capability inventory: The agent is instructed to read and write to the user's codebase to implement the UI and event capture logic. 4. Sanitization: The skill focuses on structural implementation and property mapping rather than content sanitization of the survey definition data.
Audit Metadata