pp-discord

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

• Third-party content exposure detected (medium risk: 0.30). Source: SKILL.md’s required runtime workflow uses discord-pp-cli recall "<user's question>" --agent and then reads/schedules “candidates”, “playbook.notes”, and “playbook.steps” derived from prior sessions based on that free-text query, meaning outsider-provided text directly influences the CLI’s internal decision inputs at runtime.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill's prerequisite/setup invokes remote code installation via go install (e.g., "go install github.com/mvanhorn/printing-press-library/library/productivity/discord/cmd/discord-pp-cli@latest" and "go install github.com/mvanhorn/printing-press-library/library/productivity/discord/cmd/discord-pp-mcp@latest"), which downloads and builds remote code as a required dependency for the skill.