The Agent Skills Directory

[COMMAND_EXECUTION]: The skill invokes a local script using node "${CLAUDE_PLUGIN_ROOT}/scripts/spec-superflow.mjs" sync <change-dir>. This is an expected use of project-specific tooling within the local environment for managing specifications.- [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface due to its core function of processing untrusted data. • Ingestion points: The skill reads markdown content from user-provided files in workflow/changes/<change-name>/specs/. • Boundary markers: No explicit delimiters or instructions are used to separate user data from agent instructions. • Capability inventory: The skill has file-write access to workflow/specs/ and command execution via the node CLI. • Sanitization: No sanitization or filtering of the input specification content is described. This allows for a theoretical risk where instructions embedded in a specification file could influence the agent's behavior during the sync process.

spec-syncer