The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The scripts/install-skill-seekers.sh script clones a third-party repository from a personal account (yusufkaraaslan/Skill_Seekers) that is not identified as a trusted vendor or well-known service.\n- [REMOTE_CODE_EXECUTION]: The skill installs external Python packages using pip from the downloaded repository and executes scripts contained within it to perform documentation scraping.\n- [COMMAND_EXECUTION]: The SKILL.md and reference files (e.g., references/skill-seekers-integration.md) instruct the agent to run multiple shell commands, including bash, git, and python3, to manage the creation process.\n- [PROMPT_INJECTION]: The skill exposes an [INDIRECT_PROMPT_INJECTION] surface as it is designed to scrape untrusted external websites and documents and use their content to generate instructions for AI agents.\n
Ingestion points: User-specified URLs, GitHub repositories, and PDF files.\n
Boundary markers: No boundary markers or instructions to ignore embedded commands are included in the processing pipeline.\n
Capability inventory: The agent can write to the local filesystem and execute subprocesses.\n
Sanitization: Scraped content is not sanitized or validated before being incorporated into the generated SKILL.md files.

skill-factory