fatigue-frequency-manager
Pass
Audited by Gen Agent Trust Hub on Sep 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external campaign and analytics data from CSV, GA4, and ecommerce exports (SKILL.md). While this creates an ingestion surface for untrusted data, the skill includes explicit security instructions for the agent to treat these files as untrusted input and to avoid executing instructions potentially embedded within them.
- Ingestion points: Ad platform CSV reports, GA4 traffic exports, and ecommerce store revenue exports.
- Boundary markers: Explicit instructions are present, warning the agent to never execute instructions embedded in data and to use exported values only as data.
- Capability inventory: Uses a local Python script (ledger.py) for data processing; no direct network exfiltration or system modification capabilities are listed outside of data logging.
- Sanitization: The agent is instructed to manually verify the source of the data and use time-series values for delta calculations rather than raw execution.
- [DYNAMIC_EXECUTION]: The skill utilizes a local utility script, ledger.py, to manage and analyze time-series data for trending.
- Execution pattern: The skill uses a subprocess call via python3 to execute a script located within the plugin's connectors directory (scripts/connectors/ledger.py).
- Purpose: The script records and computes trend slopes for advertising metrics like CTR and CVR to diagnose audience saturation.
Audit Metadata