literature-review-tools

Pass

Audited by Gen Agent Trust Hub on Sep 4, 2026

Risk Level: SAFECOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSDYNAMIC_EXECUTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The launcher script scripts/litrun.py facilitates the installation of external Python packages from PyPI and clones code repositories from GitHub (e.g., gpt-researcher and storm) to launch web interfaces. While these targets are well-known research projects, the capability to download and execute remote code is a significant attack surface.
  • [COMMAND_EXECUTION]: The skill makes extensive use of subprocess.run in scripts/litrun.py to drive external command-line tools for PDF extraction, research synthesis, and environment management. It passes user-provided arguments through the launcher to these external processes.
  • [DYNAMIC_EXECUTION]: The litrun.py script dynamically constructs and executes shell commands for tool management. Specifically, it can execute Python code snippets via python -c (in cmd_run for libraries) and launches long-running UI servers using uvicorn or streamlit based on configurations in recipes/recipes.json.
  • [INDIRECT_PROMPT_INJECTION]: The skill fetches content (titles, abstracts, and potentially full texts) from multiple external academic databases such as OpenAlex, PubMed, and arXiv.
  • Ingestion points: Data enters the agent's context through scripts/fetch_papers.py, scripts/fetch_arxiv.py, scripts/fetch_openalex.py, scripts/fetch_pubmed.py, and scripts/resolve_oa.py.
  • Boundary markers: The skill lacks explicit boundary markers or instructions to the LLM to ignore potentially malicious commands embedded within retrieved research data.
  • Capability inventory: The skill possesses capabilities to write files, make network requests, and execute subprocesses through the litrun.py launcher.
  • Sanitization: The scripts perform minimal sanitization, such as stripping HTML tags from abstracts, but do not filter for adversarial prompt content.
  • [EXTERNAL_DOWNLOADS]: The skill performs numerous downloads from external sources, including academic API endpoints (e.g., api.openalex.org, eutils.ncbi.nlm.nih.gov) and file downloads from repositories like arXiv and Europe PMC. It also clones repositories from GitHub for UI-based research tools.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 4, 2026, 01:08 AM
Security Audit — agent-trust-hub — literature-review-tools