The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill fetches and processes content from Canvas, such as assignment descriptions, syllabus details, and module pages, which are authored by external users.
Ingestion points: External data is ingested through subcommands like assignments view, syllabus, pages, and discussions in SKILL.md.
Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded commands within the retrieved text.
Capability inventory: The skill allows the execution of shell commands via the canvaslms CLI tool.
Sanitization: There is no mention of sanitizing or escaping the content retrieved from Canvas before it is presented to or processed by the agent.
[COMMAND_EXECUTION]: The skill generates shell commands for the canvaslms CLI. User-provided strings, such as course codes and assignment names, are interpolated as arguments for these commands.
[EXTERNAL_DOWNLOADS]: The skill requires the installation of the canvaslms and cryptography Python packages from PyPI. The source code for the CLI tool is hosted on GitHub at github.com/dbosk/canvaslms.

canvas