cloud

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt explicitly instructs the agent to accept a user-provided SHIPLIGHT_API_TOKEN and append it into the project's .env file (i.e., include the secret value in generated output/actions), which requires handling/outputting the secret verbatim.

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly fetches user-authored content from the Shiplight cloud API (e.g., GET /v1/test-cases, /v1/functions, /test-case-results and /v1/s3/file) and the SKILL.md workflows require the agent to parse/convert and act on that content (convert cloud URLs, save YAML, download artifacts, trigger/poll runs), so untrusted third-party test cases/functions/artifacts could inject instructions that influence subsequent actions.