Features

Features for reviewable AI coding loops

Hal gives coding agents a workflow: plan the requirement, bound the context, inject local standards, run implementation, and preserve the trail for human review.

01 //

Plan the work before the agent writes code

Turn product intent into stories, acceptance criteria, and reviewable state before an AI coding agent touches the repo.

02 //

Fresh context for every implementation story

Keep long AI coding work from turning into one drifting chat by giving each story its own bounded context window.

03 //

Bring your coding agent into the Hal loop

Use Codex, Claude Code, or Pi inside the same PRD-driven loop without rewriting your planning workflow.

04 //

Inject project standards into each agent loop

Give each AI coding iteration the repo-specific commands, conventions, and review rules it needs to stay consistent.

05 //

Preserve the state behind every coding loop

Preserve progress, reports, workflow files, and loop state so AI coding work can be paused, inspected, and resumed.

06 //

A deterministic pipeline for AI coding work

Move through planning, conversion, validation, implementation, reporting, and archive behavior as one explicit coding loop.

07 //

Hal vs direct agent prompting

Compare Hal's PRD-native coding loop with direct AI coding agent prompting for planning, validation, reviewability, and merge safety.

Run the first loop Inspect source