github-trending
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes repository names and descriptions fetched from GitHub (untrusted external sources). These are interpolated into an Obsidian Markdown note. While the
md_escapefunction provides basic sanitization against Markdown table breakage, the ingested text could potentially contain malicious instructions intended to influence the agent if it later reads the generated note. - Ingestion points:
fetch_trending.py(GitHub Trending HTML) anddiscover_projects.py(GitHub API JSON). - Boundary markers: The generated Markdown does not use explicit boundary markers or instructions to treat the repository data as untrusted content.
- Capability inventory: The skill has file-writing capabilities through
write_trending_note.pyto the user's Obsidian vault. - Sanitization: Employs
md_escapeto handle pipe characters and newlines in repository descriptions.- [COMMAND_EXECUTION]: The skill relies on executing local Python scripts (fetch_trending.py,discover_projects.py,write_trending_note.py) to perform data retrieval and processing. These scripts are provided within the skill package and do not appear to execute arbitrary user-supplied commands.
Audit Metadata