developer-growth-analysis

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill explicitly reads the user's local chat history (including pastedContents) and requires "evidence-based" observations in the generated report and delivery to Slack, which creates a high risk the model will output secret values (API keys/auth configs) verbatim from the history.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.75). Outsider-authored free text is ingested via the runtime HackerNews search/curation step (“Find Learning Resources” using Rube MCP), where article titles/descriptions/comments/discussion text from public posts are added into the report that the agent then sends to the LLM context.