The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external content from files located in the .agent-work/tasks/ directory, which may include tickets, research, or other upstream artifacts.
Ingestion points: The agent reads all relevant files in the task directory in full (documented in SKILL.md).
Boundary markers: Instructions do not specify the use of delimiters or ignore-instructions when processing external content.
Capability inventory: The skill uses directory listing (ls), file read/write operations, and specialized research tools like codebase-analyzer and web-search-researcher provided by the host environment.
Sanitization: The skill lacks explicit instructions to sanitize or validate the content of ingested files before they are incorporated into the PRD.

create-prd