The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill fetches and summarizes content from numerous external RSS feeds listed in references/rss.txt. This creates a surface for indirect prompt injection where malicious instructions embedded in a news article or feed summary could influence the agent's behavior during report generation.
Ingestion points: RSS feed summaries and descriptions retrieved using the mcp_rss_get_feed tool.
Capability inventory: Data normalization via scripts/report.py, summarization by the agent, and output delivery via open-webui-skill.
Boundary markers: None identified; the skill does not use specific delimiters or instructions to treat the ingested feed data as untrusted text.
Sanitization: No explicit content filtering or sanitization is performed on the feed data before processing.
[EXTERNAL_DOWNLOADS]: The skill maintains a registry of external RSS feed URLs in references/rss.txt. It actively retrieves data from these remote endpoints to provide the most recent news. The list includes several well-known technology companies and services.
[DYNAMIC_EXECUTION]: The test script tests/test_tech_news_report_skill.py dynamically loads the scripts/report.py module using standard library functions for local module execution. This procedure uses static local file paths and does not process untrusted code from external sources.

tech-news-report-skill