The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes and presents output from external command executions to the agent, which could contain instructions that influence the agent's behavior.
Ingestion points: The skill reads the output of openspec status and openspec instructions within SKILL.md (Steps 4 and 5) to determine artifact paths, next steps, and templates.
Boundary markers: There are no explicit delimiters or instructions provided to the agent to treat the command output as untrusted data or to ignore embedded instructions.
Capability inventory: The skill is authorized to execute openspec commands via the Bash(openspec:*) toolset.
Sanitization: No sanitization or validation of the CLI output is performed before it is integrated into the agent's context.
[COMMAND_EXECUTION]: The skill uses the openspec CLI to manage workflows, including creating new changes, listing stores, and viewing artifact instructions. These operations are scoped to the openspec binary as defined in the allowed-tools frontmatter.

openspec-new-change