The Agent Skills Directory

[COMMAND_EXECUTION]: The skill defines a workflow involving the execution of TypeScript modules via a shell environment and system cron. This is the intended functionality for managing automated pipelines.
[EXTERNAL_DOWNLOADS]: Mentions the installation and use of the pi-pipeliner npm package as a dependency for the framework.
[INDIRECT_PROMPT_INJECTION]: The skill creates an attack surface by instructing the agent to read configuration and API details from local files (~/repos/pipeliner/README.md) and to generate executable modules.
Ingestion points: ~/repos/pipeliner/README.md (SKILL.md)
Boundary markers: Absent. The agent is instructed to read the file directly for API details.
Capability inventory: Shell execution, npm package interaction, and invocation of the claude CLI tool.
Sanitization: No specific sanitization or validation logic is mentioned for the content read from the repository documentation.

pipeliner