fetch-abstract-to-kb

Pass

Audited by Gen Agent Trust Hub on Mar 10, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The script executes the openclaw CLI tool using the subprocess module to manage browser sessions and extract page content. Arguments are passed as a list, preventing shell injection vulnerabilities.
  • [EXTERNAL_DOWNLOADS]: The skill navigates to external URLs (specifically doi.org and its redirects) via the OpenClaw browser to retrieve metadata for scholarly articles.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests untrusted content from the web and writes it to a database.
  • Ingestion points: Data is fetched from external websites through the extract_abstract_from_doi function in scripts/fetch_abstract_to_kb.py.
  • Boundary markers: No delimiters or instructions are used to isolate the fetched text from other agent inputs.
  • Capability inventory: The skill possesses the ability to execute CLI commands and perform database write operations.
  • Sanitization: The script normalizes whitespace but does not filter for potential instructions or prompt injection patterns in the retrieved text.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 10, 2026, 03:55 AM