The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes external instructions and templates retrieved from the openspec CLI tool, which are used to guide the agent's behavior and file generation.
Ingestion points: The JSON output from commands like openspec instructions and openspec status is used to define artifact content and build order.
Boundary markers: The skill provides instructions to the agent to treat fields like context and rules as internal constraints rather than including them in the final output artifacts.
Capability inventory: The skill utilizes the Bash tool to run the openspec CLI and write files to the local file system.
Sanitization: No explicit sanitization logic is provided for the data returned by the CLI tool before it is interpreted by the agent.
[COMMAND_EXECUTION]: The skill constructs shell commands dynamically using variables such as change names provided by the user and store or artifact IDs discovered during the workflow.
Evidence: The skill defines command templates like openspec new change "<name>" and openspec status --change "<name>" --json --store "<id>". These executions are restricted to the openspec CLI by the allowed-tools configuration.

openspec-propose