web-qa-bot

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The prompt includes plaintext credentials in example test steps (e.g., "text: testpass"), which encourages embedding secrets directly in test definitions and would require the agent to output those secret values verbatim.

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

• Third-party content exposure detected (high risk: 1.00). The skill's CLI and code (e.g., web-qa-bot smoke in SKILL.md / src/cli.ts and the smokeTest/goto/snapshot flows in src/smoke.ts and src/browser.ts) navigate to arbitrary user-supplied public URLs and parse accessibility snapshots and console output from those pages, which the agent reads and interprets to make testing decisions and drive interactions.