The Agent Skills Directory

[DATA_EXPOSURE]: The skill instructs the agent to read local documentation files (e.g., docs/domains/registry.md, docs/domains/domain-map.md) if they exist. These are project-specific paths used to gather architectural context for the research prompt. No access to sensitive system directories or environment variables is requested.- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from local project files. While this creates a surface for indirect prompt injection if those files were maliciously crafted, the risk is mitigated by the skill's lack of dangerous capabilities. • Ingestion points: docs/domains/registry.md, docs/domains/domain-map.md, and docs/domains//domain.md (SKILL.md) • Boundary markers: None present • Capability inventory: No tools, file-write, network, or command execution capabilities detected • Sanitization: None present.

deepresearch-v2