The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data from project documentation files, creating a surface for indirect prompt injection attacks.
Ingestion points: The agent reads PRD.md, PLAN.md, and SPEC.md from the project directory to generate tasks.
Boundary markers: The instructions do not specify the use of delimiters or 'ignore embedded instructions' warnings when processing these files.
Capability inventory: The skill has file-read access to documentation and file-write access to the task ledger (project/TASKS.md).
Sanitization: No sanitization, escaping, or validation of the input file content is mentioned prior to task generation.

decompose