The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: Fetches the sumble-pp-cli tool using npx from the @mvanhorn NPM registry and go install from the vendor's GitHub repository (github.com/mvanhorn/printing-press-library).- [COMMAND_EXECUTION]: The skill uses the Bash tool to execute sumble-pp-cli commands. It also instructs the user to run installation commands (npx, go install) and MCP registration commands.- [DATA_EXFILTRATION]: The sumble-pp-cli tool supports a --deliver webhook:<url> flag that allows routing command output (which may contain enriched organizational or personal data) to an external URL. It also features a feedback mechanism that can be configured to send local data to an external endpoint via SUMBLE_FEEDBACK_ENDPOINT.- [PROMPT_INJECTION]: The skill processes user-provided input through the $ARGUMENTS variable and interpolates it directly into shell commands (sumble-pp-cli <command> [subcommand] [args] --agent). This creates a surface for indirect prompt injection where malicious input could influence command execution.
Ingestion points: User-provided $ARGUMENTS in SKILL.md.
Boundary markers: Absent; input is interpolated directly into shell commands.
Capability inventory: Uses the Bash tool to execute commands.
Sanitization: Absent; the agent is instructed to match the query and execute without explicit validation or escaping.

pp-sumble