The Agent Skills Directory

[COMMAND_EXECUTION]: The skill invokes the openspec CLI tool to perform actions such as listing changes, checking status, and creating new change proposals. These commands are executed through a restricted Bash tool interface.
[SAFE]: The skill includes explicit guardrails that forbid implementation or code writing, ensuring the agent remains in an analytical and exploratory role rather than an operational one.
[SAFE]: Analysis of the skill instructions and metadata revealed no evidence of obfuscation, data exfiltration, or persistence mechanisms. The interaction with the local file system is restricted to the project root and OpenSpec-related directories, which is consistent with its stated purpose.

openspec-explore