auto-verify

Automate task verification by applying predefined criteria patterns to drive tests across HTTP, browser, and file-based checks, reducing manual review. The system orchestrates curl, bash, and browser tools to generate, execute, and interpret commands with structured results during verify-task processes.

Yes, automated criteria checks require predefined patterns stored in a PATTERNS.md file. These patterns detect verification criteria across HTTP, browser, file, and environment checks to drive the automated tests and generate structured results.

The best way to verify HTTP and browser-based criteria automatically is using a pattern-driven approach that orchestrates curl, bash, and browser tools with timeout handling. This coordinates command execution and interprets results during phase-checkpoint workflows.

Yes, you can fall back to manual review when automated verification fails. The system attempts automated criteria verification during verify-task or phase-checkpoint steps and preserves manual review as a fallback mechanism for cases that cannot be automatically verified.

Pattern-driven verification handles file and environment checks by detecting criteria patterns and using available tools like bash to execute commands. It interprets the command outputs to generate structured results with timeout handling for reliable automated testing.

Use automated verification instead of manual task checks when validating repetitive criteria in AI task pipelines to accelerate workstreams. Automated verification applies across HTTP, browser, and file checks, falling back to manual review only when necessary.