ara-compiler
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process a wide variety of untrusted external inputs (PDFs, GitHub repos, logs) without instructions for sanitization or the use of boundary markers.\n
- Ingestion points: The SKILL.md file explicitly mentions reading PDF papers, GitHub repositories, code files, and experiment logs.\n
- Boundary markers: There are no requirements for delimiters or warnings to ignore instructions found within the input data.\n
- Capability inventory: The instructions direct the agent to use Bash, Read, Write, Edit, Glob, and Grep tools directly.\n
- Sanitization: No sanitization, escaping, or filtering of external content is described.\n- [COMMAND_EXECUTION]: The skill mandates the use of the Bash tool for environment exploration and artifact validation.\n- [EXTERNAL_DOWNLOADS]: The agent is instructed to fetch content from URLs and clone GitHub repositories provided through user-supplied arguments.
Audit Metadata