The Agent Skills Directory

[PRIVILEGE_ESCALATION]: The skill instructs the agent to use sudo apt-get install to install system-level packages including ffmpeg, wget, and unzip on Debian/Ubuntu systems.
[EXTERNAL_DOWNLOADS]: The workflow involves running a repository script (download.sh) which fetches large external binary resources, specifically model weights and assets, from remote servers.
[COMMAND_EXECUTION]: The skill performs extensive shell command execution to manage Python environments (venv, conda), install dependencies via pip, and launch long-running network services using uvicorn and custom Python modules.
[INDIRECT_PROMPT_INJECTION]:
Ingestion points: The agent is directed to read the README.md and config.toml files from the local repository checkout to determine project state.
Boundary markers: The instructions lack boundary markers or warnings to disregard instructions embedded within these external files.
Capability inventory: The skill possesses significant capabilities including git clone, sudo usage, pip installation, shell script execution, and the ability to start listening web servers.
Sanitization: Content from the ingested files is not sanitized or validated before being processed by the agent's context.

openstoryline-install