The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the Bash tool to execute shell commands for verifying name availability and managing local files.
It executes curl to retrieve HTTP status codes from external registries.
It utilizes the GitHub CLI (gh) to check user handle availability and list repository names.
It uses git commands (worktree list, ls-tree, fetch) to determine serial numbers for generating documentation file names.
[EXTERNAL_DOWNLOADS]: The skill interacts with external services to verify project names.
It queries rdap.org for domain registration status.
It queries the official npm registry at registry.npmjs.org for package availability.
It performs web searches and fetches content from top results to check for existing products or trademarks.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from external or user-provided sources, which presents an attack surface for indirect prompt injection.
Ingestion points: Untrusted data enters the agent context through user-provided project descriptions, naming convention examples, and repository names fetched via gh repo list. It also ingests content from web search results.
Boundary markers: The instructions do not specify the use of delimiters or specific safety instructions to isolate ingested content from the primary prompt instructions.
Capability inventory: The skill has access to Bash (shell command execution), Read, and Write (file system operations restricted to the docs/names/ directory).
Sanitization: While the skill applies specific rubrics and transliteration rules for name generation, it lacks general sanitization for the text ingested from external sources before it is processed by the agent.

namekit