The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to read and analyze codebase files to help the user explore architecture, which introduces a surface for indirect instructions if those files contain malicious content.
Ingestion points: Local codebase files, project glossaries, and change artifacts (e.g., proposal.md, design.md).
Boundary markers: The instructions do not specify the use of delimiters or specific warnings to ignore instructions embedded within the files being read.
Capability inventory: The agent can execute openspec CLI commands and write markdown artifacts to specific project directories.
Sanitization: There are no explicit instructions for sanitizing or validating the content retrieved from the codebase before it is used to generate summaries or new artifacts.
[COMMAND_EXECUTION]: The skill executes the openspec CLI (openspec list --json) to gather context about the project. This is a legitimate use of the primary tool associated with the skill's vendor.

openspec-explore