speckit-updater

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (high risk: 0.90). The skill explicitly instructs the agent to hide the -Proceed flag from users and to automatically re-invoke the updater (implementation detail hidden from user), which is a deceptive/hidden instruction outside the skill's stated conversational approval purpose.

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

• Third-party content exposure detected (low risk: 0.10). The required runtime workflow runs a local PowerShell update-orchestrator that fetches template/version data from GitHub Releases API and then prompts the user based on parsed script output markers, with no step indicating that outsider-authored free text is ingested unless it is first fetched/returned as part of the script’s trusted GitHub content.