Agent skill · software engineering · alirezarezvani
agent-harness
Use when an agent should drive a goal in one of this repo's domains to a verified close, before starting the work. Compiles the goal into a task plan with goal_compiler.py and runs it through loop_controller.py, which verifies each task, caps retries and escalates.
Why this skill is useful
Adds a comprehensive framework for automating task execution and verification across multiple domains, including scripts for goal compilation and loop control that enhance the agent's capabilities.
What it needs
Requires python3 installed locally. About 4k tokens when loaded. Last updated 2026-08-05. 23,939 stars on the source repository.
What this skill does
Agent Harness You are a harness operator, not a hero. The loop — not your optimism — decides when work is done. Your job: compile the goal into tasks with checks, execute one task at a time, let the controller adjudicate verification, and stop when the state machine says stop. The contract Three layers, all JSON: a committed per-domain manifest (what skills/tools/checks exist), a per-goal plan (which tasks, which verifications, what "done" means), and a per-run state file (the single source of truth; a fresh session resumes from it alone). Quick start Regenerate a manifest after skills change (diff-stable, CI-checkable): Hard rules 1. Never adjudicate your own verification. verify runs the checks via subprocess; a passing record --phase verify without --evidence is rejected (exit 6). You do not get to declare a task verified. 2. Never modify a gate you are judged by. Check commands come from the manifest/plan. Editing a check to make it pass is the reward-hacking failure mode (see references/verificationdiscipline.md) — same invariant as autoresearch-agent's locked evaluator. 3. One task at a time, writes serialized. Parallelize reading and judging, never two tasks writing the same artifact (references/agenticloopcanon.md). 4. Retry means a changed approach. Same command + same input = same failure. The retry directive says so; honor it. 5. Budgets are terminal states, not suggestions. maxattemptspertask → escalated (exit 2); maxloopiterations → escalate (exit 5). Exhausted budgets are never reported as success — a human waives (close --waive T3 --reason "..."), you don't. 6. Fresh context beats long context. Every next directive is executable by a new session reading only the plan + state files. Long-running goals: run each iteration as its own session against the durable state. 7. State lives in .agent-harness/ — never in .agenthub/, .autoresearch/, or docs/TC/ (those belong to sibling skills). 8. Plan and state files are a trust boundary. …
How to use it
Reference it in AdaL, Claude Code, Cursor or any coding agent — nothing to install:
@skills alirezarezvani/agent-harness--dd8e59