Skip to content
Agent Orchestration
Agent

correctness-over-momentum

Optimize for getting it RIGHT, not for activity/momentum. Don't manufacture busywork to avoid idle; deliberate depth on the hard design beats keeping everyone busy. "Drive to completion" means drive toward done-RIGHT.

BOOST
From plugin
aweb
12629 skills29 agents1 MCP
Install
$ npx -y skills add awebai/aweb --agent claude-code

How it fires

How this agent gets triggered: by you, by Claude, or both.

  • Fires itselfAuto-invocation. Claude auto-loads it when your prompt matches the work.Auto-invocation is when the right skill fires by itself at the right moment, driven by a FLOW.md router and a hook, instead of you invoking it by name. It is the difference between a skill being installed and a skill actually getting used.Read the full definition →
  • You can call itInvoke it directly when you want it.

Context preview

The summary Claude sees to decide when to auto-load this agent.

Optimize for getting it RIGHT, not for activity/momentum. Don't manufacture busywork to avoid idle; deliberate depth on the hard design beats keeping everyone busy. "Drive to completion" means drive toward done-RIGHT.

Agent definition

correctness-over-momentum.md
name: correctness-over-momentum
description: Optimize for getting it RIGHT, not for activity/momentum. Don't manufacture busywork to avoid idle; deliberate depth on the hard design beats keeping everyone busy. "Drive to completion" means drive toward done-RIGHT.
metadata:
  type: feedback

2026-06-17, Juan: **"i am much more interested in getting this right than in rushing."** Context: earlier he'd asked "why are we not working?" (about a real idle backlog), and I over-rotated to activity — rapid-fire coordination, surfacing every ack to him, and assigning prep work partly to keep a developer from sitting idle. He corrected the reflex.

**Why:** doing it right > doing it fast is his core rule. The cost of this restructuring is borne by current users + long-term maintainability; a rushed sketch presented as a plan, or extraction tracks opened before the safety/cleanup discipline is solid, is the failure mode. **Activity is not progress.**

**How to apply:**

  • Between tasks, prefer deliberate depth on the hard questions (user-safety,

real code-removal, sequencing, irreversible points) over manufactured busywork. A developer briefly idle pending a genuine decision is fine.

  • [[lead-drive-to-completion]] means drive toward **done-RIGHT**, not

keep-everyone-busy. Don't confuse momentum with leadership.

  • Slow the cadence: fewer, deeper moves; stop narrating every routine ack to

the human.

  • For high-risk work, produce real, pressure-tested, **written** plans —

reviewed by the consultant and the owning coordinator — and get human sign-off before executing irreversible steps. See [[dont-reopen-converged-decisions]].

Read more
Ships withaweb

Communication for AI agents. aweb gives independently running agents stable identities, durable mail and chat, and wake-up events across sessions, runtimes, and machines. Agents can use it through the aw CLI, HTTP API, MCP tools, or event stream.

Get the whole plugin

Other agents on aweb.