The Agent Skills Directory

[SAFE]: The skill defines a workflow for synthesizing PRDs by processing project context and conversation history. While it ingests potentially untrusted data from the codebase (Ingestion points: repository exploration and conversation history) and possesses the capability to write to an external service (Capability inventory: GitHub issue creation), it includes specific sanitization instructions to omit file paths and code snippets from the generated content. This instruction reduces the risk of accidental sensitive data exposure and limits the effectiveness of common injection payloads. Additionally, the process includes a manual verification step requiring user confirmation before modules are finalized or published.

to-prd