The Agent Skills Directory

[COMMAND_EXECUTION]: The skill performs filesystem operations using shell commands such as mkdir -p and mv. These commands use variables (planningHome.changesDir, changeRoot, and name) that are dynamically retrieved from the output of the openspec CLI tool. While the instructions use double quotes to wrap these variables, the integrity of the operation depends on the paths provided by the external tool.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from external, potentially untrusted sources to determine its execution flow.
Ingestion points: Data is ingested from the JSON output of openspec status --json and the contents of a local tasks.md file.
Boundary markers: The skill parses structured JSON for CLI output, which provides some implicit structure, but does not specify explicit delimiters or sanitization for the content of the tasks.md file.
Capability inventory: The skill has the capability to execute shell commands (mkdir, mv), read files, and invoke other agent tools/skills (e.g., openspec-sync-specs).
Sanitization: Shell variables are double-quoted to prevent basic word splitting and globbing, and the skill uses the AskUserQuestion tool to seek user confirmation before proceeding with archives that have incomplete artifacts or tasks.

openspec-archive-change