The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill facilitates the installation of the conductor-pp-cli tool and an MCP server through npx and go install from vendor-controlled repositories. Evidence includes commands targeting @mvanhorn/printing-press-library and github.com/mvanhorn/printing-press-library.
[DYNAMIC_EXECUTION]: The skill implements a 'learning loop' where the agent retrieves and executes 'playbooks' (JSON-defined command sequences) from a local store based on historical interactions. This allows the execution of dynamically synthesized command paths.
[DATA_EXFILTRATION]: The CLI tool supports a --deliver webhook:<url> flag that POSTs command output to an external URL. This can be used to transmit sensitive session transcripts or authentication metadata to arbitrary third-party endpoints.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied queries through a recall and teach loop, creating an attack surface where malicious input could poison the local learning store with adversarial 'playbooks' that influence future agent behavior. Evidence chain: Ingestion points include user questions in the 'Automatic Learning' section; boundary markers are absent as playbooks are treated as 'ground truth'; capabilities include full CLI control and shell access; sanitization relies on manual PII stripping.
[COMMAND_EXECUTION]: The skill enables comprehensive management of cloud workspaces and sessions through shell commands, including a tool for running SQL queries against session data.

pp-conductor