The Agent Skills Directory

External Package Execution: The skill uses npx to execute the openspec command-line tool. This pattern involves downloading and running code from the public npm registry at runtime, which is a common practice in development workflows but represents a dependency on an external package registry.
Command Execution: The instructions involve executing shell commands (e.g., npx openspec new, npx openspec status) to manage project artifacts. Users should ensure these commands are executed within a controlled environment.
Indirect Prompt Injection Surface: The skill processes external data to generate new content, which presents a surface for indirect prompt injection.
Ingestion points: Reads JSON output from npx openspec instructions and content from existing dependency files.
Boundary markers: The skill includes instructions to the agent to treat external context and rules as constraints rather than content to be copied into output files.
Capability inventory: The skill has the ability to execute shell commands via the openspec CLI and write new files to the filesystem.
Sanitization: There are no explicit sanitization or validation steps described for the data ingested from external artifacts.

openspec-ff-change