The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted data from external sources provided by the user (emails, chat logs, documents) to populate sections of business alignment documents and contracts.
Ingestion points: Identified in step-01c-offer-extract.md and step-01d-extract-info.md, where the user is prompted to share communications for extraction.
Boundary markers: The instructions do not specify the use of delimiters or 'ignore embedded instructions' warnings when processing this external data.
Capability inventory: The skill facilitates file writing operations, specifically creating docs/wds-1-project-brief/pitch.md, contract.md, and signoff.md in the project workspace.
Sanitization: The instructions lack explicit guidelines for sanitizing, escaping, or filtering the content extracted from external communications.

wds-0-alignment-signoff