The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to Indirect Prompt Injection because it ingests data from external sources to generate executable HTML/JS code.
Ingestion points: Reads text and metadata from user-provided .ppt and .pptx files using python-pptx as described in SKILL.md.
Boundary markers: Absent. There are no instructions to the agent to treat extracted slide content as untrusted data or to use delimiters.
Capability inventory: The skill generates .html files containing JavaScript and uses shell commands (open, start, xdg-open) to execute them. It also runs python3 scripts for extraction.
Sanitization: Absent. The skill does not provide instructions to escape or sanitize the content extracted from slides before inserting it into the final presentation code.
[COMMAND_EXECUTION]: The skill utilizes shell commands to interact with the operating system for its primary functionality.
Evidence: Uses open, xdg-open, and start to launch the generated presentation in the default browser.
Evidence: Spawns python3 processes to extract content from PowerPoint files.
[EXTERNAL_DOWNLOADS]: The skill requires external software to perform file conversions.
Evidence: Instructs the agent to check for and potentially install the python-pptx library from the Python Package Index (PyPI).

frontend-slides