auditing-experiments-flags
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from PostHog experiment and feature flag configurations (names, descriptions, variant keys). Maliciously crafted data in these fields could attempt to influence the agent's audit report or subsequent actions.
- Ingestion points: Data is ingested through tools including experiment-get, experiment-list, feature-flag-get-definition, feature-flag-get-all, and feature-flags-activity-retrieve (SKILL.md).
- Boundary markers: Absent; the instructions do not specify delimiters or warnings to ignore instructions embedded within the experiment or flag data.
- Capability inventory: The skill generates markdown reports for the user and has the ability to create and write to notebooks using project tools (SKILL.md).
- Sanitization: Absent; the skill interprets and reports string values from the API (such as hypotheses in descriptions) without explicit sanitization or escaping.
- Mitigation: Wrap external content in delimiters with explicit 'ignore embedded instructions' warnings and sanitize/escape external content before interpolating into output.
Audit Metadata