render-debug
Fail
Audited by Gen Agent Trust Hub on Aug 1, 2026
Risk Level: HIGHPROMPT_INJECTIONDATA_EXFILTRATIONCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: Deceptive metadata in SKILL.md claims the author is 'Render', which is inconsistent with the provided author 'Sarthib7'. This impersonation of the service provider could mislead users or agents regarding the skill's origin and safety.
- [DATA_EXFILTRATION]: The skill exposes sensitive information by providing tools to retrieve application logs via list_logs and database content via query_render_postgres.
- [COMMAND_EXECUTION]: The skill uses tools to execute arbitrary SQL queries via query_render_postgres and invokes the Render CLI for tasks such as retrieving service lists, logs, and deployment status.
- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection as it ingests untrusted log and database data without sanitization or boundary markers. Ingestion points: list_logs and query_render_postgres found in SKILL.md and references/database-debugging.md. Boundary markers: Absent. Capability inventory: update_environment_variables, query_render_postgres, and list_deploys. Sanitization: Absent.
Recommendations
- AI detected serious security threats
Audit Metadata