Track persistent goals with external evaluation for agents until tasks are done.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "goal-engine" yet — see the docs or source repo.
Use goal-engine to create a goal for my CLI agent: 'Make all project tests pass.' After each run, evaluate the test results, record progress, and keep updating the code until all tests pass or a stop condition is reached.
A persistently tracked goal workflow with per-run evaluations, progress state, and automatic continuation until the condition is met.
Set a deployment goal: 'Confirm the service is deployed successfully and health checks pass.' Let the agent execute commands across multiple turns, evaluate outcomes externally, and continue troubleshooting until the condition is satisfied.
An automated loop centered on deployment success criteria that preserves state and keeps progressing until validation is complete.
Create a goal: 'Collect and organize key materials on a topic until a defined list of subquestions is covered.' After each retrieval and synthesis step, use external rules to evaluate coverage and decide whether another turn is needed.
A persistent research goal record showing completed subquestions, current coverage, and whether execution should continue.
Turn vague intentions into concrete, measurable goals and clear success criteria.
Design, audit, and optimize AI coding agent loops and orchestration systems.
Optimize content for visibility across ChatGPT, Perplexity, Claude, and Gemini.
Give coding agents memory, validation, and feedback across sessions.
Run AI agents through OpenAI-compatible APIs with memory and multi-step workflows.
Turn plain-English goals into verified, looped, observable IDE agent build runs.