The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute system shell commands (open on macOS or xdg-open on Linux) to display generated HTML validation reports to the user.
[PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external data from the user's codebase and existing documentation, which creates a surface for indirect prompt injection.
Ingestion points: The skill reads repository source code, existing PRD drafts, and environment configuration files from the project directory (e.g., in SKILL.md and references/validate.md).
Boundary markers: There are no specific instructions to wrap external content in delimiters or include explicit warnings to the agent to ignore instructions embedded in that data.
Capability inventory: The skill has the ability to write files to the workspace, perform network requests via vendor tools, and execute shell commands to open local files.
Sanitization: The instructions do not detail any sanitization or validation of the content extracted from project files before it is used in the synthesis of new documents.
[EXTERNAL_DOWNLOADS]: The skill fetches the latest product catalog and documentation from Juspay's official domains (*.juspay.io, *.juspay.in) and its authorized CloudFront distribution. This is intended behavior for grounding the documentation in real-world product specifications.

jp-prd