The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from task descriptions and project files, which could contain instructions designed to manipulate the agent.
Ingestion points: task-description argument and repository files such as CLAUDE.md, design/gdd/, and production/sprints/ (found in SKILL.md).
Boundary markers: The skill does not employ delimiters or instructions to ignore commands within the processed data.
Capability inventory: The skill is restricted to Read, Glob, and Grep tools and lacks network, file-write, or command execution capabilities.
Sanitization: No filtering or validation is applied to the ingested content.

estimate