The Agent Skills Directory

[Command Execution]: The skill utilizes standard shell commands such as mkdir and mv to organize the project directory. These operations are restricted to the local file system within the openspec/ directory for archiving purposes.
[Tool Integration]: It executes the openspec utility via npx to interact with the project's workflow state. This allows the agent to list changes and verify the status of artifacts before proceeding with archival.
[Data Ingestion Surface]: The skill processes information from the local environment, including tasks.md and CLI output, to assess project completion.
Ingestion points: Reads the contents of tasks.md and parses JSON status from the openspec CLI.
Boundary markers: No explicit data delimiters are used, but the skill relies on structured JSON parsing for tool status.
Capability inventory: Includes directory creation (mkdir) and moving files (mv).
Sanitization: Uses structured status data to drive logic, though it does not explicitly mention sanitization of the markdown task file content.
[User Confirmation Guardrails]: The skill is designed to prevent unintended actions by requiring the user to explicitly select changes and confirm operations if incomplete tasks or artifacts are detected.

openspec-archive-change