gi

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly instructs collecting API keys/URL and passing them as --apikey/--api-key command arguments and extracting apikey from /gi inputs, which requires the agent to include secret values verbatim in generated commands or code.

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

• Third-party content exposure detected (high risk: 0.80). The helper explicitly calls an arbitrary API base URL (apiBaseUrl from /gi-setup or --url) and reads/streams the endpoint's JSON/events (scripts/generate-image.mjs callImageEndpoint/readStreamResponse) and even fetches returned image URLs (saveGeneratedImage uses fetch(image.value)), so untrusted third-party responses can be ingested and can influence retry/fallback logic and subsequent tool actions.