testing-dags
Pass
Audited by Gen Agent Trust Hub on Apr 30, 2026
Risk Level: SAFEREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill uses
uvxto execute theastro-airflow-mcppackage. This is a tool provided by the vendor 'astronomer' for management and debugging operations. - [COMMAND_EXECUTION]: The skill instructs the agent to run CLI commands such as
afandastro devto manage Airflow workflows. These are expected behaviors for the skill's primary purpose. - [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface.
- Ingestion points: The agent retrieves task logs via
af tasks logs(SKILL.md), which may contain content from external or untrusted sources. - Boundary markers: No specific delimiters or instructions are used to separate log data from agent instructions.
- Capability inventory: The agent has the capability to execute shell commands and modify file content (e.g., DAG files and
requirements.txt) as described in the Phase 3 'Fix and Retest' section of SKILL.md. - Sanitization: There is no evidence of log content validation or sanitization before it is consumed by the agent.
Audit Metadata