The Agent Skills Directory

[SAFE]: The skill consists of documentation and TypeScript code snippets for interacting with the official Grammarly Enterprise API. All network requests target legitimate placeholders (e.g., GRAMMARLY_API) or well-known Grammarly domains for documentation.
[INDIRECT_PROMPT_INJECTION]: The skill defines functions for managing team roles and memberships that ingest user-provided data such as email addresses and team names. While these are potential entry points for indirect injection, they are presented as standard administrative templates for the stated purpose of the skill.
Ingestion points: email, team, and role parameters in the assignTeamRole and revokeTeamAccess functions located in SKILL.md.
Boundary markers: None present in the code snippets.
Capability inventory: The skill frontmatter specifies access to Read, Write, and Grep tools, and the code demonstrates network communication capabilities via the fetch API.
Sanitization: None implemented in the provided conceptual code templates.

grammarly-enterprise-rbac