skills/openclaw/skills/paper-fetcher/Gen Agent Trust Hub

paper-fetcher

Pass

Audited by Gen Agent Trust Hub on Mar 18, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill performs network requests to fetch HTML data and download PDF files from the external domain sci-hub.su.
  • [COMMAND_EXECUTION]: The script scripts/fetch_paper.py executes system commands by calling the curl binary via Python's subprocess.run to facilitate downloads.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface due to the ingestion of untrusted external content.
  • Ingestion points: Processes user-supplied DOIs and remote HTML content fetched from the Sci-Hub website within scripts/fetch_paper.py.
  • Boundary markers: The script lacks explicit markers or instructions to isolate or ignore potential instructions embedded within the fetched HTML content during parsing.
  • Capability inventory: The skill possesses the capability to create directories and write files to the workspace through Python's pathlib and curl.
  • Sanitization: The implementation includes logic to clean DOI prefixes and replace directory traversal characters in filenames with underscores.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 18, 2026, 10:34 AM