The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill requires installing the crates-io-pp-cli and crates-io-pp-mcp binaries via npx (@mvanhorn/printing-press-library) and go install (github.com/mvanhorn/printing-press-library/...). These tools are provided by the skill's author.
[DATA_EXFILTRATION]: The CLI tool supports a --deliver webhook:<url> flag, allowing output to be sent to any external URL via HTTP POST, which could be misused to exfiltrate processed data.
[DATA_EXFILTRATION]: The CLI tool supports a --deliver file:<path> flag, enabling output to be written to arbitrary file paths, creating a risk of unauthorized file writes or overwriting sensitive system files.
[COMMAND_EXECUTION]: The skill directs the agent to execute the crates-io-pp-cli binary with arguments provided through user input.
[PROMPT_INJECTION]: The skill processes external data from crates.io, creating an indirect prompt injection surface. Ingestion points: Crate metadata and search results fetched from crates.io. Boundary markers: No delimiters or instructions are provided to the agent to ignore instructions within the fetched data. Capability inventory: The skill allows file-writing and network POST operations through the delivery sinks. Sanitization: No sanitization or validation of the external content is described.

pp-crates-io