firstsun-harness
Installation
SKILL.md
Forked from walkinglabs/learn-harness-engineering on 2026-07-04 and now maintained independently in this repo. It no longer tracks upstream changes.
firstsun-harness
Use this skill to make a repository easier for coding agents to start, stay in scope, verify work, and resume across sessions. Keep the harness small enough that agents actually follow it.
Not for model selection, prompt tuning in isolation, chat UI design, or general app architecture.
Core Model
Every useful coding-agent harness has five subsystems:
| Subsystem | Minimal artifact | Purpose |
|---|---|---|
| Instructions | AGENTS.md or CLAUDE.md |
Startup path, working rules, definition of done |
| State | feature_list.json, progress.md, archive/YYYY-MM.md |
Current feature, status, evidence, next step; finished work archived monthly |
| Verification | init.sh or documented commands |
Tests/checks the agent must run before claiming done |
| Scope | Feature dependencies and done criteria | Prevents overreach and half-finished work |
| Lifecycle | end-of-session routine, session-handoff.md (gitignored per-checkout scratch) |
Makes the next session restartable without merge conflicts across worktrees |