finding-deleted-feature-flags
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses a local Python script
scripts/strip_deleted_suffix.pyto process feature flag keys. The script is static and performs simple string manipulation on provided JSON input, presenting no risk of arbitrary execution. - [INDIRECT_PROMPT_INJECTION]: The skill processes data from external database and API sources, creating a potential attack surface for indirect injection.
- Ingestion points: Data is ingested from the
system.feature_flagstable and theposthog:feature-flags-activity-retrieveAPI responses (File: SKILL.md). - Boundary markers: No specific delimiters are used to isolate external data within the agent's context.
- Capability inventory: The skill possesses capabilities for SQL execution, API interaction, and local script execution.
- Sanitization: The skill uses a deterministic string manipulation script to clean feature flag keys before they are reported to the user.
Audit Metadata