The Agent Skills Directory

[EXTERNAL_DOWNLOADS] (HIGH): The skill instructs the agent to run pnpm create fragno@latest. This downloads and executes code from a third-party registry and an author that is not on the trusted sources list.
[REMOTE_CODE_EXECUTION] (HIGH): The use of pnpm create involves executing arbitrary initialization scripts from a remote package, posing a significant security risk if the package is compromised.
[PROMPT_INJECTION] (HIGH): The skill is highly vulnerable to Indirect Prompt Injection. It fetches markdown documentation and API data at runtime via curl from https://fragno.dev.
Ingestion points: Documentation URLs in SKILL.md and references/ directory used as source material for code generation.
Boundary markers: None present; the agent is instructed to read and follow these external guides directly as ground truth.
Capability inventory: Full project scaffolding, code generation, and package installation capabilities.
Sanitization: No validation or sanitization of the fetched content is performed before processing.
[COMMAND_EXECUTION] (MEDIUM): The skill makes extensive use of shell commands (pnpm, curl) which could be abused if the agent is tricked by malicious external content into executing unintended parameters.

fragno-fragment-creation