The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands to run a local generator script and open the resulting HTML file.
Evidence: SKILL.md contains instructions for the agent to execute node scripts/generate-cf-deck.mjs and open presentations/<name>/index.html.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input (pasted content) which is saved to a file and later processed by the script.
Ingestion points: User input is received in Step 0 and written to skills/examples/<name>.md in Step 1.
Boundary markers: Absent. The agent is not explicitly instructed to use delimiters or ignore embedded instructions within the user-provided content.
Capability inventory: The skill uses fs (via the script) to read/write files and shell commands to execute the generation process.
Sanitization: The generator script in scripts/generate-cf-deck.mjs implements an escapeHtml function to encode HTML entities before they are inserted into the slide templates, mitigating potential injection risks in the generated output.

slide-generator