account-activity
Pass
Audited by Gen Agent Trust Hub on Sep 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data from blockchain transaction history and account positions.
- Ingestion points: Data is ingested through tools like
get_user_activityandget_user_positions(SKILL.md). - Boundary markers: The instructions do not define explicit delimiters or warnings to ignore instructions embedded within the transaction data.
- Capability inventory: The skill instructions focus on data retrieval and reporting; no high-risk capabilities like subprocess execution or file system writes are present in the provided file.
- Sanitization: No evidence of sanitization or validation of the retrieved blockchain data is provided in the instructions.
Audit Metadata