The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes shell commands (mkdir -p and mv) to organize and archive directories. It interpolates the <name> variable, which can be provided by the user or inferred from conversation, directly into these shell strings. If the agent does not sanitize this input, it could lead to the execution of arbitrary commands if the name contains shell metacharacters like ;, &&, or backticks.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external files and tool outputs to determine its workflow logic.
Ingestion points: Reads JSON output from openspec list and openspec status, and parses the contents of tasks.md and delta spec files in openspec/changes/<name>/specs/.
Boundary markers: The instructions do not define delimiters or specific warnings to the agent to ignore instructions that might be embedded within the task descriptions or spec files.
Capability inventory: The skill has the ability to move files (mv), create directories (mkdir), and trigger a subagent (via the Task tool) to execute the openspec-sync-specs skill.
Sanitization: There is no explicit requirement to sanitize or validate the content of the tasks or spec files before processing them or passing summaries to the subagent.

openspec-archive-change