The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to perform autonomous research by searching and synthesizing information from the web. This ingestion of untrusted external data presents a potential surface for indirect prompt injection, where malicious instructions hidden in research sources could attempt to influence the agent's behavior.
Ingestion points: The scripts/research.py script (referenced in SKILL.md) ingest content from external websites during research tasks.
Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded commands in the retrieved data.
Capability inventory: The skill uses the httpx library to perform network operations to fetch data.
Sanitization: No explicit sanitization or filtering of the retrieved research content is documented.
[EXTERNAL_DOWNLOADS]: The skill requires the installation of the httpx Python package via a local requirements.txt file. httpx is a well-known, standard library for HTTP requests.

deep-research