The Agent Skills Directory

[COMMAND_EXECUTION]: Employs the @prismatic-io/prism CLI tool for project initialization, metadata inspection, and local development or testing tasks as described in SKILL.md.
[EXTERNAL_DOWNLOADS]: Fetches configuration and examples from official Prismatic documentation (prismatic.io) and the vendor's public GitHub repositories.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process external data sources such as API specifications and service documentation to generate connector code.
Ingestion points: API specifications for service contracts and official documentation referenced in SKILL.md.
Boundary markers: The skill relies on the agent's ability to distinguish between instructions and data while following official contracts, and instructs the agent to 'Verify claims about unsupported capabilities against official sources'.
Capability inventory: Execution of shell commands via prism CLI (init, dev, get, search) and runtime code execution through the Spectral testing harness (SKILL.md).
Sanitization: Instructions direct the agent to verify all external claims against official sources and to validate implementations through automated tests (Spectral harness) and manual inspection before publication.

prismatic-components