The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the openspec CLI (specifically list, status, and instructions commands) to manage the workflow. These commands are local and consistent with the skill's stated purpose of implementing tasks via the OpenSpec toolset.
[DATA_EXPOSURE]: The skill reads local project files listed in the contextFiles output of the CLI to provide the agent with necessary context for code implementation. It also modifies task tracking files to mark completion. These operations are restricted to the local workspace and follow standard developer tool patterns.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from local files (project context and task descriptions) which represents a potential attack surface. However, the skill includes explicit guardrails such as pausing for user clarification if tasks are ambiguous and keeping changes minimal and scoped, which mitigates the risk of unintended behavior from malicious file content.

openspec-apply-change