The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill instructs the user/agent to install the keenable-pp-cli tool and MCP server using npx -y @mvanhorn/printing-press-library and go install github.com/mvanhorn/printing-press-library/.... This involves downloading and executing code from external repositories.
[DYNAMIC_EXECUTION]: The skill implements an 'Automatic learning' loop where the agent fetches 'playbooks' from a local database via the recall command. These playbooks contain structured lists of shell commands (steps) that the agent is explicitly instructed to execute in order. This creates a mechanism for executing dynamically generated or previously stored command sequences based on natural language triggers.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the web via fetch and web-search commands. It also processes 'notes' and 'playbooks' from its learning store. Malicious content in these sources could potentially influence the agent's behavior or poison the learning loop via the teach or playbook amend commands.
Ingestion points: keenable-pp-cli fetch, keenable-pp-cli web-search, and keenable-pp-cli recall.
Boundary markers: None explicitly mentioned for separating web content or learned notes from agent instructions.
Capability inventory: Shell command execution via keenable-pp-cli and Read Bash.
Sanitization: The skill recommends stripping personal identifiers (PII) from queries before teaching, but does not specify sanitization or validation of the fetched web content or command steps themselves.
[DATA_EXFILTRATION]: The CLI supports a --deliver webhook:<url> flag, which allows the agent to POST the output of any command to an arbitrary remote URL. Additionally, a feedback command is included that can send local data to a remote endpoint if configured.
[COMMAND_EXECUTION]: The skill is built around executing the keenable-pp-cli binary with various arguments and flags to manage web research and the local learning store.

pp-keenable