posthog-growth-workspace
Pass
Audited by Gen Agent Trust Hub on Jul 30, 2026
Risk Level: SAFE
Full Analysis
- [UNVERIFIABLE_DEPENDENCIES_AND_REMOTE_CODE_EXECUTION]: The documentation mentions using
npx @posthog/wizard@latestfor optional self-driving features. This involves executing code from a well-known service provider via the npm registry. Following the trust scope rule, this is considered a safe and standard integration with a recognized platform. - [INDIRECT_PROMPT_INJECTION]: The skill ingests data from the PostHog API (such as event names, properties, and recordings), which could potentially contain malicious instructions from an external source. The skill has capabilities to write files to the
.growth/directory and perform network operations. Mandatory Evidence Chain: 1. Ingestion points: PostHog event data and session recordings viascripts/pg-query.mjs. 2. Boundary markers: None identified. 3. Capability inventory: File writing to the.growth/directory, execution of local query scripts, and networking with the PostHog API. 4. Sanitization: No explicit sanitization of ingested API data is implemented. This is a baseline risk inherent to data-processing skills and is assessed as safe in this context.
Audit Metadata