sn-search-social-en
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill fetches and processes content from external social media platforms, creating a vulnerability surface where malicious instructions could be embedded in search results.
- Ingestion points: Data is retrieved from external APIs in
scripts/reddit_search.py,scripts/twitter_search.py, andscripts/youtube_search.pyand subsequently processed by the agent. - Boundary markers: While the output is structured as JSON, there are no specific instructions or delimiters defined to warn the agent to ignore potential commands embedded within the fetched content.
- Capability inventory: The skill's capabilities are limited to performing network GET requests and printing results to standard output. No file writing, system command execution, or dynamic code evaluation was detected in the scripts.
- Sanitization: The scripts perform basic character truncation on the retrieved text but do not implement sanitization or filtering to detect or neutralize prompt injection markers.
Audit Metadata