generate-theories
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external scientific literature which could potentially contain malicious instructions.
- Ingestion points: User-provided research questions and content from external papers retrieved via the asta tool.
- Boundary markers: No specific delimiters or boundary instructions are defined to separate untrusted data from agent instructions.
- Capability inventory: The skill uses the Bash tool to execute CLI commands and manage local files.
- Sanitization: The instructions do not define sanitization procedures for data extracted from papers before it is incorporated into the theory generation process.
- [COMMAND_EXECUTION]: The skill utilizes the Bash tool to interact with the asta command-line interface. It constructs shell commands using user-supplied research questions and context passed verbatim to the tool. Commands are used for authentication, theory generation, and artifact management.
- [EXTERNAL_DOWNLOADS]: The skill references the vendor's GitHub repository for the internal Theorizer project. It points to https://github.com/allenai/asta-theorizer-internal for technical documentation and agent specifications.
Audit Metadata