The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The script scripts/view_pypi_latest_version.sh retrieves metadata from the official PyPI registry and pipes the response directly to a Python interpreter to parse the version string.
[EXTERNAL_DOWNLOADS]: The skill instructions in SKILL.md direct the agent to perform a git clone of the agentscope-ai/agentscope repository from GitHub into the local environment to facilitate exploration of documentation and code examples.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface (Category 8) because it is designed to ingest and process data from an external repository while possessing code execution capabilities.
Ingestion points: Content from the repository cloned from github.com/agentscope-ai/agentscope (Step 1 in SKILL.md).
Boundary markers: Absent; the instructions do not specify the use of delimiters or warnings when the agent reads the cloned repository files.
Capability inventory: The skill provides and documents tools for execute_shell_command and execute_python_code (examples in SKILL.md and references/multi_agent_orchestration.md).
Sanitization: None; the agent is encouraged to read and adapt examples from the external repository directly.
[COMMAND_EXECUTION]: The skill includes a Python script scripts/view_module_signature.py that utilizes the inspect module to dynamically examine the structure and signatures of the agentscope library at runtime.

agentscope-skill