The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied text (Instagram captions and search keywords) which are analyzed by a local script and the agent. This represents an indirect prompt injection surface.
Ingestion points: SKILL.md and scripts/caption.py process user-provided caption text and keywords.
Boundary markers: Absent. The instructions do not specify the use of delimiters or warnings to ignore instructions within the processed text.
Capability inventory: The agent can execute the local Python script (scripts/caption.py) and interact with Instagram-related tools (list_platforms, ig-reel).
Sanitization: Absent. The text is processed directly for character counting and pattern matching without sanitization of potential command structures.
[COMMAND_EXECUTION]: The skill instructions direct the agent to execute a local Python script (scripts/caption.py) to perform caption linting and search term verification. This script is part of the skill package and uses standard Python libraries for processing.

ig-caption