wiki-enrich

Pass

Audited by Gen Agent Trust Hub on Oct 27, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes local Python scripts (research_wiki.py) and standard system utilities like curl, grep, and git. These are used for managing the wiki state and fetching data from official academic repositories.
  • [EXTERNAL_DOWNLOADS]: Fetches research paper abstracts and overviews from well-known and reputable services including alphaxiv.org and arxiv.org. These operations are consistent with the skill's stated purpose of enriching a research wiki.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from external sources (arXiv and alphaXiv) to generate summaries. This creates a surface for indirect prompt injection if an attacker-controlled paper contains malicious instructions designed to influence the agent's behavior during the enrichment process.
  • Ingestion points: WebFetch from alphaxiv.org, curl from export.arxiv.org, and reading local markdown files.
  • Boundary markers: None explicitly defined to separate untrusted source text from the agent's internal summarization instructions.
  • Capability inventory: Uses Edit and Write tools to modify local filesystem content and Bash for utility operations.
  • Sanitization: No specific sanitization or filtering of the fetched content for prompt injection markers is documented.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 27, 2026, 03:07 AM