The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes project-specific CLI tools, specifically openspec status and openspec validate, as part of its workspace verification process. These tools are scoped to project validation tasks within the local environment.
[PROMPT_INJECTION]: The skill instructions require the agent to read SKILL.md files from the local workspace to verify native skill prerequisites. This creates a surface for indirect prompt injection where a malicious file within the project directory could attempt to influence the agent's logic.
Ingestion points: SKILL.md files discovered within the project root directory during the native skill verification phase.
Boundary markers: None identified; the skill does not specify delimiters or instructions to ignore embedded commands within the ingested files.
Capability inventory: The agent is permitted to navigate the local file system and execute specific openspec CLI commands.
Sanitization: There is no evidence of sanitization or content validation for the data retrieved from the SKILL.md files.

openspec-workspace-gates