The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes shell commands (mkdir, mv) and the openspec CLI to manage file system operations. It uses double quotes for variable interpolation (e.g., "<changeRoot>", "<name>") which mitigates standard shell injection risks.
[EXTERNAL_DOWNLOADS]: The skill requires the openspec CLI to be available on the system. As this is a tool associated with the vendor, it is considered a legitimate dependency for the skill's functionality.
[PROMPT_INJECTION]: The skill processes untrusted data which presents an indirect prompt injection surface.
Ingestion points: The skill reads the contents of tasks.md and parses JSON output from openspec list and openspec status commands (SKILL.md).
Boundary markers: No explicit delimiters or boundary markers are defined for the interpolated data.
Capability inventory: The skill has the ability to execute shell commands and invoke other agent tools.
Sanitization: Shell commands use quoting. Critically, the skill includes human-in-the-loop checkpoints using the AskUserQuestion tool to confirm actions whenever incomplete tasks or artifacts are detected in the processed data.

openspec-archive-change