The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes multiple shell commands and scripts to perform its tasks.
It uses curl to interact with the Open edX Studio API at $STUDIO_URL.
It executes an inline Python script using a heredoc to parse and display course outline data from a JSON file.
It invokes a preflight authentication check by calling a Node.js script located at a specific file path within the agent's skills directory (.claude/skills/iblai-api-studio-auth/scripts/studio-login.mjs).
[INDIRECT_PROMPT_INJECTION]: The skill processes course structure data that is fetched from an external API, creating a surface for potential instruction injection.
Ingestion points: Course outline data is fetched from the Studio API via GET /xblock/outline/{ROOT} and saved to a local file outline.json in SKILL.md.
Boundary markers: The instructions do not define boundary markers or provide explicit directions to the agent to disregard potential instructions embedded within the fetched course metadata.
Capability inventory: The skill has the capability to execute shell commands (curl), perform network requests, and execute Python scripts, which could be leveraged if an injection is successful.
Sanitization: The included Python script extracts display_name and visibility_state fields from the JSON data and prints them to the console without any sanitization or validation, allowing raw external strings to enter the agent's context.

iblai-api-studio-publish