The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill instructs the user to install a binary using npx -y @mvanhorn/printing-press-library or go install github.com/mvanhorn/printing-press-library/.... This involves downloading and executing code from external repositories controlled by the vendor.
[DATA_EXFILTRATION]: The CLI tool supports a --deliver webhook:<url> flag, which allows routing the output of any command to an arbitrary external URL. This capability could be exploited to exfiltrate sensitive data or credentials if the agent is prompted to send command results to a malicious endpoint.
[DATA_EXFILTRATION]: The feedback command can be configured to automatically POST locally recorded notes to an external endpoint via the PUSHOVER_FEEDBACK_ENDPOINT environment variable.
[INDIRECT_PROMPT_INJECTION]: The skill includes commands to sync and download messages (inbox sync, messages download) from the Pushover service. These messages are ingested into the agent's context and could contain malicious instructions.
Ingestion points: pushover-pp-cli inbox sync, pushover-pp-cli messages download.
Boundary markers: None specified in the instructions for separating ingested message content from the system prompt.
Capability inventory: File writing (local SQLite store), network operations (Pushover API, webhooks), and command execution.
Sanitization: Not explicitly mentioned; the skill processes raw message data from the service.

pp-pushover