The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill includes patterns that ingest external data (e.g., $(cat $file) in data_processing.sh) and pipe it directly into AI model prompts without sanitization or boundary markers.
Ingestion points: data_processing.sh reads all .txt files in a directory.
Boundary markers: Absent. Content is interpolated directly into a JSON prompt string.
Capability inventory: The skill uses belt app run to execute various AI models and curl for alerting.
Sanitization: None detected.
[EXTERNAL_DOWNLOADS]: The documentation includes references to external scripts and installation guides from GitHub.
Evidence: https://raw.githubusercontent.com/inference-sh/skills/refs/heads/main/cli-install.md is referenced for installation instructions.
[COMMAND_EXECUTION]: The skill utilizes the Bash tool and provides multiple shell script templates that execute the belt CLI tool, curl, and file system operations.
Evidence: Extensive use of belt app run, mkdir, and redirection in provided automation patterns.

ai-automation-workflows