The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses uvx and uv run to execute shell commands for web searching and content retrieval.
[EXTERNAL_DOWNLOADS]: Dynamically installs and runs the web-forager and ddgs packages from public registries at runtime to perform research tasks.
[EXTERNAL_DOWNLOADS]: Fetches web content through r.jina.ai, an external well-known service used to convert HTML pages into LLM-friendly markdown.
[PROMPT_INJECTION]: Presents an indirect prompt injection surface (Category 8) because it ingests untrusted data from the web.
Ingestion points: Content fetched from arbitrary URLs via web-forager or jina.ai as described in SKILL.md.
Boundary markers: Absent; the instructions do not include specific delimiters or warnings to ignore instructions found within fetched web pages.
Capability inventory: Subprocess execution via uvx and uv run, and network operations via curl.
Sanitization: None; the skill relies on the agent's internal safety guardrails to process external content safely.

deep-research