affaan-m/
ECC
affaan-m/ECCHarnesses
The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
258k
Roadmap and handoff workflow for Claude Code and Codex
Foreman keeps a roadmap in the project so your agent can decide what comes next without losing the reason, constraints, or open blockers. It checks the roadmap against the code, then turns finished work into evidence that can inform later tasks.
34 stars•5 forks•Updated 7d ago
Who it's for
Builders who use Claude Code or Codex and want each session to pick up from the last one with project context intact.
What it delivers
You can resume work with a clear next task, its reason, and its checked context instead of re-explaining the project.
What it does
Project roadmap beside the code
Stores the plan in the repository so task order and context travel with the project.
Next-task recommendation
Answers "what's next?" with a suggested task and the reason behind it.
Grounded plan check
Checks stale plans against the current code before the agent acts.
Task status tracking
Shows status, blockers, and work waiting for acceptance.
Checked handoff prompts
Creates a prompt for the next step without requiring a roadmap entry.
Separate acceptance from test success
Keeps acceptance of finished work distinct from a passing run.
README
Foreman
Your plan stays next to your code. The next task arrives with its context checked.
Available on
Get started · What is this? · How it works · What you can do · Evidence
What is this?
You finish a session with a fix, a side request and an issue still waiting. Next time, you need to know which one comes first and why. Foreman keeps that context in a roadmap beside your code.
Ask "what’s next?" and it recommends a task with its reason and a prompt checked against the project’s files. Completed work brings evidence back into the plan; acceptance stays a separate decision.
Why you'd want it
- Pick up where the last session stopped.
- Keep the task, its reason and its constraints together.
- Check stale plans against the code before acting.
- Keep acceptance of finished work separate from a successful test run.
How it works
The roadmap lives in your project. Foreman reads it, checks the relevant context and helps you choose what to do next. Optional lessons let later tasks learn from earlier work without creating a separate knowledge base.
What you can do
| You say | You get |
|---|---|
| Add this to the roadmap | Requested work recorded with its reason and boundaries |
| Where are we? | Status, blockers and work waiting for acceptance |
| What's next? | A recommended task and a choice of where to run it |
| Check whether the plan still matches the code | A grounded review before choosing work |
| Craft a prompt for this | A checked handoff without requiring a roadmap entry |
Get started
Choose the assistant you use. Its edition page has the installation steps, commands and compatibility notes for your setup.
| Your assistant | Status | Next step |
|---|---|---|
| Claude Code | Available | Install and get started |
| Codex | Available | Install and get started |
Good to know
Foreman is for a solo developer. It does not become a team tracker, code-review service, unattended scheduler or workflow server. Your project keeps its own data. The optional ledger is off until requested; disabling it deletes no existing notes.
Evidence and benchmarks
The task trail above illustrates the workflow. The recorded demo below comes from Claude Code; it is not a Codex measurement. Evidence and methodology.
Watch the recorded Claude Code demo
Measurements belong to the model and setup that produced them. Each edition keeps its own results, limitations and any measurements still missing:
Going deeper
Research and validation · Benchmark instruments and retained evidence · Foundry
License
MIT — see LICENSE.
Files in the repo
Repository payload•4 top-level entries
- .github
- assets
- LICENSE
- README.md
Discussion (0)
Ask about usage, or say what you built with itSign in to join the discussion.
No comments yet. Be the first to say what this is good for.
More harnesses
from vibe coding to agentic engineering - practice makes claude perfect
66k
ruvnet/rufloHarnesses
🌊 The original agent meta-harness. Deploy intelligent multi-player swarms, coordinate autonomous workflows, and build conversational AI systems. Features adaptive memory, self-learning intelligence, RAG integration, and native Claude Code / Codex / Hermes and many more Integrated
72k
cobusgreyling/loop-engineeringHarnesses
Practical patterns, starters & CLI tools for loop engineering with AI coding agents. Design systems that prompt and orchestrate agents (inspired by Addy Osmani and Boris Cherny). Includes loop-audit, loop-init, loop-cost.
11k
open-gsd/
gsd-core
open-gsd/gsd-coreHarnesses
Git. Ship. Done - Core
9.3k
1jehuang/jcodeHarnesses
The most RAM efficient harness
19k