The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the vendor's own command-line utility, openspec list --json, to retrieve project state and metadata for context.
[PROMPT_INJECTION]: The skill is designed to read and process external markdown artifacts such as proposal.md, design.md, and spec.md from the local openspec/ directory. This provides a surface for indirect prompt injection if these files originate from an untrusted source.
Ingestion points: reads from openspec/changes/<name>/proposal.md, design.md, tasks.md, and specs/<capability>/spec.md as specified in SKILL.md.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands within the processed files are present.
Capability inventory: The skill can execute openspec CLI commands and perform file read operations.
Sanitization: No explicit sanitization or filtering of the content read from external files is performed before processing.

openspec-explore