The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill requires the installation of an external third-party tool called OpenClawCLI from clawhub.ai.
Evidence: Found in the YAML frontmatter description and the Troubleshooting section pointing to https://clawhub.ai/.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from external repositories that could contain malicious instructions designed to influence the agent's behavior during previews or documentation generation.
Ingestion points: Dataset files (CSV, JSON, JSONL, Parquet, etc.) downloaded from repositories like Kaggle, Hugging Face, UCI, and Data.gov, which are then processed by the preview and datacard commands in scripts/dataset.py.
Boundary markers: The instructions lack explicit delimiters or guidance to the agent to ignore instructions embedded within the data content.
Capability inventory: The skill environment includes file system write access (via --output flags), reading of local files, and network communication for downloads.
Sanitization: No evidence of input sanitization, content filtering, or validation is provided for the data being parsed and summarized.

dataset-finder