Superpowers 6.4 Released with Agent Expansion, Revamped Inline Execution, and Diagnostics

Superpowers 6.4 Released with Agent Expansion, Revamped Inline Execution, and Diagnostics Superpowers 6.4 introduces support for three new coding agents, overhauls its native inline execution flow for modern frontier models, adds a dedicated diagnostics skill, and refines planning and code review behaviors. New support has been added for Meta's Muse, OpenCode 2.0, and Qwen Code. Native Execution (`executing-plans`) has been reinvented for modern frontier models, measuring about twice as fast and half as expensive as Subagent-Driven Development in evaluations. A new `diagnosing-superpowers` skill analyzes agent history to investigate unexpected behaviors and assist with bug reports. Refinements to `brainstorming`, `writing-plans`, and code review prompts aim to prevent step-skipping, unrequested work, and aggressive review loops. Model Support and Planning Refinements Superpowers 6.4 expands compatibility to Meta's Muse, OpenCode 2.0, and Qwen Code, adapting to the quirks of newer frontier models. The release updates `brainstorming` and `writing-plans` skills to prevent agents from skipping necessary steps or generating unrequested work. Code reviewer prompts were also adjusted to exercise reasonable judgment, helping curb overzealous models from getting trapped in endless review loops. Reinvented Native Execution and Performance The release revamps Native Execution (`executing-plans`), an inline alternative to the orchestrator-based Subagent-Driven Development. In evals with Claude Code, Native Execution performed approximately twice as fast and at half the cost of subagent setups on frontier models. While running without Superpowers remains faster and cheaper, testing alongside Astra and Fable demonstrated that builds lacking Superpowers implementation skills were significantly more bug-prone. Diagnostics and Availability The update adds `diagnosing-superpowers`, a tool designed to inspect coding agent session histories during unexpected outcomes and assist users in preparing bug reports. Superpowers 6.4 is available immediately via the Superpowers Marketplace and OpenAI Codex, while Anthropic and other platform marketplaces update according to their respective release schedules.

Superpowers 6.4 发布：重构原生执行并新增三款编程代理支持 Superpowers 6.4 正式发布，新增对 Meta Muse、OpenCode 2.0 和 Qwen Code 的支持，重构了原生内联执行能力，并加入诊断技能及审查调优。 新增对 Meta 的 Muse、OpenCode 2.0 以及 Qwen Code 三款编程代理的支持。 重构面向前沿模型的原生内联执行技能，其速度约为子代理驱动开发的两倍，成本降低一半。 改进计划与审查提示词以减少跳步和过度循环，并推出用于排查异常的诊断技能。 通过 Superpowers Marketplace 或 OpenAI Codex 安装的用户现已可用，Anthropic 等平台依自身排期更新。 新代理支持与规划优化 6.4 版本针对前沿模型的行为特点进行了适配，新增对 Meta Muse、OpenCode 2.0 和 Qwen Code 的支持。同时，更新了头脑风暴与规划编写技能，防止代理跳过步骤或执行未要求的冗余工作。 原生执行重构与性能评估 针对更强的前沿模型，6.4 重构了原生内联执行技能。基准评估显示，搭配前沿模型时，原生执行的速度约为子代理驱动开发的两倍，成本仅为其一半。测试表明，完全脱离 Superpowers 构建虽成本更低更快，但在 Astra 和 Fable 等模型下产生的 Bug 明显更多。 诊断技能与审查提示词调优 新版本引入了 diagnosing-superpowers 技能，用于深入审查代理历史记录并协助生成 Bug 报告。此外，调优了代码审查提示词，引入更务实的判断标准，以避免过度审查导致代理陷入死循环。

I'm pleased to announce the release of Superpowers 6.4. This release adds new support for three coding agents: Meta's Muse, OpenCode 2.0 and Qwen Code. Every new model behaves just a bit differently than the ones before it. As the frontier models get more capable, they also develop new quirks. A lot of the work in this release is meant to help the latest models use Superpowers effectively. We've done some work on `brainstorming` and `writing-plans` to help steer your agents away from skipping steps or doing significant unrequested work. Superpowers offers two plan execution styles. One is "Subagent-Driven Development", which uses the primary agent session as an orchestrator, dispatching all implementation and test tasks to right-sized subagents with limited context and limited mandates. The other is "Native (inline) Execution", which dates to the earliest days of Superpowers. It was originally intended for a human-in-the-loop flow where you acted as a go-between for two agentic sessions. As the frontier models have improved, they've gotten better at executing work inline without getting distracted or losing the plot. Superpowers 6.4 reinvents Native Execution (the `executing-plans` skill) for today's more capable models. We've been running extensive evals comparing Subagent-Driven Development, Native Execution and "bare metal" implementations with Claude Code. With a frontier model, Native Execution is about twice as fast and half as expensive as Subagent-Driven Development. Building without Superpowers gets you results faster and cheaper, but in our testing, even with Astra and Fable, builds without Superpowers implementation skills are significantly buggier than builds using Superpowers. We are, of course, working hard to make Superpowers faster and cheaper to use. We're currently testing out some fairly big changes to the planning skills that we hope to debut soon. 6.4 also introduces a new `diagnosing-superpowers` skill, which you can use when a session has done something you didn't expect. It knows how to dig into your coding agent's history to understand what Superpowers was doing and can help you put together a good bug report. We've done a little bit of work to tune the code reviewer prompts to apply a bit more "what would a reasonable person think?" judgment. This should help overzealous models calm down and converge rather than work themselves into aggressive review loops. You can read the full release notes at [https://github.com/obra/superpowers/blob/main/RELEASE-NOTES.md](https://github.com/obra/superpowers/blob/main/RELEASE-NOTES.md) If you install Superpowers from the Superpowers Marketplace ([https://github.com/obra/superpowers-marketplace](https://github.com/obra/superpowers-marketplace)) or from OpenAI Codex, 6.4 is available in your coding agent today. Anthropic and other providers update their plugin marketplaces on their own schedules.