The Agent Skills Directory

[PROMPT_INJECTION]: The skill contains multiple surfaces for indirect prompt injection.
Ingestion points: The conditional_workflow.sh script ingests command-line arguments, and data_processing.sh reads the entire content of external text files.
Boundary markers: No delimiters or explicit instructions to ignore embedded commands are used when interpolating this data into prompts.
Capability inventory: The skill can execute various AI models via the infsh tool and perform network requests via curl.
Sanitization: There is no evidence of validation or escaping of external content before it is processed by AI models.
[EXTERNAL_DOWNLOADS]: The documentation references an external installation script for the CLI tool hosted on the vendor's repository: https://raw.githubusercontent.com/inference-sh/skills/refs/heads/main/cli-install.md.
[COMMAND_EXECUTION]: The provided templates rely on executing system commands through Bash scripts and the Python subprocess module to orchestrate AI tasks.
[DATA_EXFILTRATION]: The skill includes a monitored_workflow.sh example that demonstrates sending command execution results and error logs to an external webhook URL using curl.

ai-automation-workflows