The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses uvx and uv run to execute search utilities like web-forager and ddgs. It also utilizes curl for network requests.
[EXTERNAL_DOWNLOADS]: Downloads Python packages (web-forager, ddgs, requests) from official registries at runtime to ensure tools are available.
[EXTERNAL_DOWNLOADS]: Connects to r.jina.ai (Jina Reader) to fetch and convert web pages into a format suitable for the agent.
[PROMPT_INJECTION]: The skill is designed to fetch and process arbitrary web content, which presents a surface for indirect prompt injection where malicious instructions could be embedded in the fetched articles.
Ingestion points: Web content fetched via Jina Reader or search results in SKILL.md.
Boundary markers: No explicit delimiters or instructions are provided to the agent to distinguish between the skill's instructions and the fetched data.
Capability inventory: The skill has access to shell execution (uvx, uv run, curl) and network operations.
Sanitization: No sanitization or filtering of external content is specified before the agent processes it.

fact-check