The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes content from an external source (wwdc.ai).
Ingestion points: Data is fetched from wwdc.ai (markdown, JSON) and apple.com (JSON) via curl in scripts/wwdc.sh and references/docs.md.
Boundary markers: No delimiters or instructions are used to separate external content from agent instructions.
Capability inventory: The skill has access to shell execution (scripts/wwdc.sh), network requests (curl), and Node.js for parsing.
Sanitization: No filtering or validation is performed on the content retrieved from external URLs.
[EXTERNAL_DOWNLOADS]: The skill retrieves session summaries and transcripts from external domains.
Evidence: Fetches from wwdc.ai and devimages-cdn.apple.com.
Context: These sources are associated with the vendor and a well-known service, respectively.
[COMMAND_EXECUTION]: Uses a bundled shell script to automate data retrieval and parsing.
Evidence: scripts/wwdc.sh executes curl, node, and awk commands.
Context: The script uses safe practices such as quoting variables and using node -e for parsing instead of eval.

wwdc