The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The installation script install.sh downloads and executes the uv package manager installer directly from its official distribution point at astral.sh. This is a well-known service in the Python ecosystem.
[EXTERNAL_DOWNLOADS]: The skill fetches Python dependencies from skale.dev, which is the author's verified domain, and utilizes public markdown extraction APIs from Jina.ai and markdown.new for its core functionality.
[COMMAND_EXECUTION]: The main logic in scripts/fetch.py executes local system binaries for text-based browsers including w3m, lynx, and chawan using the subprocess module with argument lists to prevent shell injection.
[PROMPT_INJECTION]: The skill's primary function involves ingesting untrusted data from arbitrary third-party URLs, which presents an attack surface for indirect prompt injection.
Ingestion points: Network data fetched in scripts/fetch.py and output from text browser subprocesses.
Boundary markers: None implemented; content is returned directly to the agent.
Capability inventory: Includes network access via the requests library and local command execution via subprocess.run across multiple files.
Sanitization: The skill identifies common error patterns in the fetched content but does not perform sanitization of text intended for downstream processing by the agent.

fetch-url