The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the openspec CLI tool through the Bash(openspec:*) capability to perform administrative and synchronization tasks, including listing stores, checking change status, and validating updated specifications.
[PROMPT_INJECTION]: The skill is subject to indirect prompt injection risks because it reads and interprets the content of "delta specs" (Markdown files) to perform intelligent merging. Maliciously crafted content within these files could attempt to influence the agent's logic during the merge process.
Ingestion points: Delta specification files located at paths returned by the openspec status command and main specification files located at <planningHome.root>/openspec/specs/.
Boundary markers: The agent relies on standard Markdown headers (e.g., ## ADDED Requirements) for parsing, but there are no explicit delimiters or "ignore instructions" warnings to isolate potentially malicious file content from the agent's operational logic.
Capability inventory: The agent has the authority to read, write, and delete files within the specification root, as well as execute allowed openspec commands.
Sanitization: The skill does not define specific sanitization or filtering steps for the text content being processed before it is integrated into the main specification files.
[SAFE]: The skill implements specific security logic to prevent directory traversal attacks during capability retirement by ensuring that file deletion does not follow symlinks to external directories.

openspec-sync-specs