The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill downloads the goodreads-pp-cli binary and associated library from github.com/mvanhorn/printing-press-library and @mvanhorn/printing-press-library. These are legitimate vendor resources belonging to the skill's author context.
[COMMAND_EXECUTION]: The skill executes shell commands using the goodreads-pp-cli and npx / go install for installation. These are necessary for the skill's primary function and follow standard developer workflows.
[DATA_EXPOSURE]: The skill manages Goodreads session cookies and GraphQL tokens via environment variables (GOODREADS_GOODREADS_COOKIE_SESSION, GOODREADS_GRAPHQL_TOKEN). It explicitly warns against leaking these in shell history and recommends secure practices like using stdin or .env files. No evidence of unauthorized exfiltration was found.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data from Goodreads (reviews, feed, shelf data) and passes it to the agent. While this presents an attack surface, the risk is inherent to the tool's purpose (a Goodreads client), and there are no signs of exploitable capabilities being combined with this data in an unsafe manner.

pp-goodreads