The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the user to use npx to fetch and execute the @seed-hypermedia/cli package. This is the standard distribution and execution method for the vendor's official command-line tools.
[REMOTE_CODE_EXECUTION]: Automated scanners flagged a command pattern involving curl piped to python3. Technical analysis of the specific command curl -sI ... | awk ... | python3 -c ... reveals it is a benign utility for URL-decoding a specific HTTP header (x-hypermedia-id). The Python code is a static string used for data transformation and does not download or execute remote scripts. The target URL belongs to the vendor's domain (hyper.media).
[COMMAND_EXECUTION]: The skill provides numerous shell commands for document lifecycle management, including creating, updating, and searching for hypermedia content. These are appropriate for a CLI-focused skill.
[INDIRECT_PROMPT_INJECTION]: The skill includes functionality for ingesting external data that could contain malicious instructions.
Ingestion points: Reads content from remote hypermedia documents, local PDF files, and JSON blocks.
Boundary markers: The instructions state that the CLI validates content structure during draft creation.
Capability inventory: The skill can perform network operations (publish to servers), read local files, and manage cryptographic keys.
Sanitization: Content is parsed from markdown or PDF into a structured JSON block tree before processing.

seed-cli