The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests external data from web pages via scripts/fetch_page.py and local files via scripts/brand_scanner.py. The SKILL.md and references/skill-contract.md files provide clear boundary markers instructing the agent to treat this data as untrusted and ignore any embedded instructions. The skill has capabilities for network access and file writing. Sanitization is present in the form of SSRF protections.
[EXTERNAL_DOWNLOADS]: scripts/fetch_page.py is a tool for fetching external web content. It includes extensive security checks to prevent Server-Side Request Forgery (SSRF), such as blocking non-global IP addresses and restricting protocols.
[DYNAMIC_EXECUTION]: scripts/citability_scorer.py uses __import__('sys') to read from standard input. This is a benign use of dynamic module loading for script compactness.

geo-llmstxt