senior-backend
Pass
Audited by Gen Agent Trust Hub on Apr 5, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONDATA_EXFILTRATION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by processing untrusted data to perform scaffolding and migration tasks.
- Ingestion points: OpenAPI YAML/JSON files (SKILL.md) and database schemas accessed via connection strings (SKILL.md).
- Boundary markers: There are no explicit delimiters or instructions to ignore embedded commands within the ingested specifications.
- Capability inventory: The skill uses scripts that can write to the local filesystem (
api_scaffolder.py), execute database operations (database_migration_tool.py), and perform network requests (api_load_tester.py). - Sanitization: No evidence of input sanitization or schema validation is provided for the contents of the ingested files.
- [COMMAND_EXECUTION]: The skill executes local Python scripts (
api_scaffolder.py,database_migration_tool.py,api_load_tester.py) via the command line. These scripts perform operations such as file scaffolding, schema analysis, and performance testing based on user-provided arguments. - [DATA_EXFILTRATION]: The
database_migration_tool.pyutility relies on database connection strings (e.g.,postgres://localhost/mydb). If a connection string for a sensitive or production database is provided, it could allow the analysis tools to access and potentially expose database schemas or data within the agent's operating environment.
Audit Metadata