The Agent Skills Directory

[COMMAND_EXECUTION]: The skill relies on the local execution of the openspec CLI. It uses subcommands such as list, status, and store list to discover project state and retrieve file paths. These commands are restricted to the local environment and the tool's specific configuration.
[INDIRECT_PROMPT_INJECTION]: The skill ingests delta specification files which are external to the skill's own code, creating an attack surface for indirect prompt injection.
Ingestion points: Processes delta spec files identified by artifactPaths.specs.existingOutputPaths as defined in SKILL.md.
Boundary markers: Parsing is limited to specific Markdown headers including ## ADDED Requirements, ## MODIFIED Requirements, and ## REMOVED Requirements.
Capability inventory: The skill possesses the ability to execute CLI commands and perform read/write operations on the local filesystem.
Sanitization: The skill requires the use of the AskUserQuestion tool to force user selection of changes and requires the agent to present a summary of all changes, which provides a layer of human-in-the-loop review.

openspec-sync-specs