ara-compiler
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted external inputs such as PDF papers, GitHub repositories, and experiment logs, which serves as a potential attack surface for indirect prompt injection.
- Ingestion points: Inputs are read using native tools (Read, Bash, Glob, Grep) as described in the
Input Reading Strategysection ofSKILL.md. - Boundary markers: The skill instructions do not specify the use of clear boundary markers or explicit commands to ignore instructions embedded within the processed research materials.
- Capability inventory: The agent has significant capabilities including
Bashfor shell command execution andWrite/Editfor file system modifications (e.g., generatingsrc/execution/Python stubs). - Sanitization: There is no evidence of sanitization, escaping, or validation of the content extracted from external research sources before it is integrated into generated artifacts.
Audit Metadata