Flow Agents

Flow Agents

Structured workflows for coding agents. Flow Agents helps Codex, Claude Code, Kiro, and CI agents keep work inspectable from idea to release readiness.

Stay on the path Turn loose requests into shaped work, plans, implementation waves, review, verification, and release decisions.
Show the evidence Keep acceptance criteria, test results, browser checks, governance reports, and gaps in durable workflow artifacts.
Use your runtime Install one bundle across Codex, Claude Code, Kiro, and local automation without rewriting the workflow for each tool.

How It Works

flowchart LR
  Idea[Idea]
  Backlog[Backlog]
  Plan[Plan]
  Build[Build]
  Review[Review]
  Verify[Verify]
  Evidence[Evidence]
  Release[Release readiness]
  Learn[Learning]

  Idea --> Backlog --> Plan --> Build --> Review --> Verify --> Evidence --> Release --> Learn
  Review -->|finding| Build
  Verify -->|gap| Build
  Evidence -->|not verified| Plan

Flow Agents adds the operating layer around the model: skills choose the right workflow, sidecars preserve state, hooks catch stop-short behavior, and evals keep the bundle honest as it changes.

Quick Start

Install into a workspace:

bash install.sh /path/to/workspace

Then ask for the workflow you want:

Use deliver for this GitHub issue. Plan it, implement it, review it, verify it, and stop if evidence is missing.

For bugs:

Use fix-bug. Reproduce the problem, diagnose root cause, implement the fix, and verify the regression path.

Use Cases

Why It Matters

Long-running agent work fails when the model loses context, skips verification, or calls partial work done. Flow Agents makes the process explicit without making the user write a perfect prompt every time. The agent gets a workflow; the developer gets artifacts they can inspect.