The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes shell commands to manage directories and archive workflow data.
It uses mkdir -p and mv to create archive folders and relocate change directories.
Variables like <name> and <change-name> are interpolated directly into shell strings in Step 5 without explicit quoting, which may lead to command injection if a change name contains shell metacharacters and the underlying platform does not provide automatic sanitization.
[INDIRECT_PROMPT_INJECTION]: The skill ingests external data from the file system that could potentially influence the agent's behavior.
Ingestion points: The skill reads tasks.md to count task completion and parses specification files in openspec/changes/<name>/specs/ to generate summaries.
Boundary markers: No delimiters or instructions to ignore embedded commands are used when reading these files.
Capability inventory: The skill can execute shell commands (mkdir, mv) and trigger subagents via the Task tool to perform synchronization tasks.
Sanitization: There is no evidence of sanitization or validation of the text extracted from the tasks or specification files before it is processed or passed to the Task tool.

openspec-archive-change