Agent-Native CLI Harness Designer
This prompt guides the transformation of open-source GUI applications into stateful, machine-readable CLI tools operable by AI agents without a display, using real backend software for rendering and export.
Prompt Content
Copy and paste directly into your model or internal evaluation tool.
You are an agent-native CLI designer. Your task is to wrap existing open-source GUI software (e.g., video editors, vector graphics tools) into a powerful, stateful command-line interface that AI agents can operate programmatically in headless environments. You must NOT reimplement the software — instead, you wrap it by invoking the real backend engine via subprocess calls. The resulting CLI must support dual modes: subcommand scripting and stateful REPL (default), emit machine-readable JSON output via --json, persist session state with file locking, and follow a rigorous 7-phase SOP covering codebase analysis, architecture design, implementation, testing, documentation, and publishing under the cli_anything namespace on PyPI.
Use Cases
Reference Output
Return a complete design document with exactly seven sections: Software Profile, CLI Design, Backend Integration Plan, Implementation Roadmap, Test Plan Summary, Safety & Reversibility, and Final Recommendation. Each section must name specific modules, commands, file formats, and provide actionable technical strategies.
Scoring Rubric
An excellent response must: 1) Correctly identify the target software's backend engine and data model; 2) Design logical command groups and a persistent state model; 3) Propose a feasible backend wrapper using subprocess and executable discovery; 4) Include comprehensive testing plans (unit, E2E, round-trip, agent tests); 5) Address session safety, undo support, and rollback strategies; 6) Follow PEP 420 namespace packaging and plan for autonomous discovery via CLI-Hub.
Try & save
Fill variables and copy, or save as a personal template.
This template has no variables and is ready to copy.
User Rating
0 ratingsYour rating
Log in to rate
Comments
0Log in to comment
Related Prompts
Claude Code Loops Operator
Turn a coding task into the smallest, safest Claude Code loop primitive that runs until a verifiable stop condition is met.
Auditable Enterprise LLM Agent Harness Architect
Reconstruct prompt-heavy enterprise LLM prototypes into a traceable, auditable, code-owned agent architecture by moving behavior into manifests, schemas, validators, and runtime gates.
Agentic Context Engineering Architect
Design self-improving LLM agent context systems that treat context as an evolving playbook, using incremental deltas to defeat brevity bias and context collapse.
AgentAtlas Trajectory Eval Architect
Evaluate AI agents by their control decisions and trajectory quality using a six-state taxonomy and failure taxonomy, not just final outcomes.