The Agent Skills Directory

[COMMAND_EXECUTION]: The skill relies on executing shell commands such as 'pnpm', 'npx', 'git', 'gh', 'tmux', and 'curl' to manage the development environment, build processes, and local server instances.
[EXTERNAL_DOWNLOADS]: Fetches dependencies via 'pnpm install' and executes the vendor-provided 'paperclipai' CLI tool via 'npx'. These resources are either well-known standard tools or originate from the skill author's own infrastructure.
[REMOTE_CODE_EXECUTION]: Employs 'eval' to dynamically source shell environment variables produced by the project's 'paperclipai' CLI tool.
[PROMPT_INJECTION]: The skill is designed to read and follow instructions from repository-hosted files including 'doc/DEVELOPING.md', 'CONTRIBUTING.md', and pull request templates. This creates an attack surface where instructions embedded in these files by external contributors could potentially influence agent behavior.
Ingestion points: 'doc/DEVELOPING.md', '.github/PULL_REQUEST_TEMPLATE.md', 'CONTRIBUTING.md'
Boundary markers: Absent
Capability inventory: Shell command execution ('git', 'pnpm', 'npx', 'gh'), file system access, and local network operations
Sanitization: Absent

paperclip-dev