The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill downloads and installs the twilio-pp-cli and twilio-pp-mcp binaries from the vendor's official GitHub repository at github.com/mvanhorn/printing-press-library using go install.
[EXTERNAL_DOWNLOADS]: The skill installs the @mvanhorn/printing-press-library package from the npm registry using npx during the setup process.
[COMMAND_EXECUTION]: The skill executes the twilio-pp-cli binary to perform Twilio operations, including creating accounts, sending messages, and initiating calls.
[DATA_EXFILTRATION]: The skill provides a --deliver webhook:<url> flag, which allows the agent to POST the output of any command—which may include sensitive communication data, call recordings, or transcriptions—to an arbitrary external URL.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data such as incoming SMS messages and call transcriptions. Malicious instructions embedded in these external sources could influence the agent's behavior during analysis or automated response workflows.
Ingestion points: Untrusted data enters the context via commands like messages-json list-message, tail-messages, recordings-json, and conversation.
Boundary markers: The instructions do not specify any boundary markers or delimiters for the agent to use when processing external communication data.
Capability inventory: The skill has extensive capabilities including file system writing (--deliver file:<path>), network requests (--deliver webhook:<url>), and the ability to modify Twilio account configurations.
Sanitization: There are no explicit sanitization or filtering steps described in the skill for handling incoming message bodies or metadata before they are processed by the agent.

pp-twilio