The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is configured to ingest and process untrusted external data, specifically contractor quotes, scope documents, and estimates. This creates a surface where malicious instructions could be embedded in these documents to influence the agent's behavior.
Ingestion points: Contractor quotes, project estimates, and scope details provided by the user for analysis and tracking in SKILL.md and projects.md.
Boundary markers: The templates provided in memory-template.md lack explicit delimiters or instructions to ignore embedded prompts within the stored user data.
Capability inventory: The agent has permissions to read and write files within the ~/home-renovation/ directory for project state management.
Sanitization: There is no evidence of sanitization or validation logic for the content of processed quotes or renovation documents.

Home Renovation