taskforge-run

仓库创建 2026年7月22日最近提交 5 天前SkillHot 收录 3 天前

这个 Skill 做什么

把 Specification 里定好的活儿按着去做出来，然后把成果交给一个完全独立、全新上下文的审查者审核；适合任务已经处于“run”准备好执行或用户要去执行/交付时用。流程是实现→独立复审→不通过就重做，审查在隔离且可审计的环境中进行，发现根本性问题会升级回 refine 或 explore。不会悄悄放宽标准或越界扩展范围，新发现的工作会拆成后续任务处理。

▸ 展开 SKILL.md 英文原文

Implement a taskforge task against its active Specification and verify it with an independent fresh-context review subagent. Use when a task's readiness is "run", or when the user asks to run, implement, execute, build, or ship a taskforge task that has a specification. Covers the implement-review-retry loop, recorded-and-auditable reviewer isolation, root-cause escalation back to taskforge-refine or taskforge-explore, and scope discipline (out-of-scope discoveries become follow-up tasks). Tasks without an active specification route to taskforge-refine instead.

Stars

Forks

仓库内 Skill

7 日增星

安装 / 使用

给你的 Agent 一句话（通用）

帮我安装这个 skill：https://raw.githubusercontent.com/hashirventhodi/taskforge-skills/main/taskforge-run/SKILL.md

或 curl 直取 SKILL.md

curl -fsSL "https://raw.githubusercontent.com/hashirventhodi/taskforge-skills/main/taskforge-run/SKILL.md"

SKILL.MD 节选查看完整文件 ↗

# taskforge-run

Implement the active Specification, then submit the work to an independent
reviewer in a fresh context. Run finishes a task or escalates it — it never
quietly lowers the bar and never expands scope.

> **Bundled vs split (docs/ADR-0001 Phase 6).** This skill bundles implement +
> review + `done` in one pass — the manual, human-driven path. Orchestrated
> flows use the split capabilities `taskforge-implement` and `taskforge-review`
> (one responsibility each; the kernel routes on the review verdict). Run
> remains fully supported; the split is the preferred path for autonomy.

**Prerequisites**: read `taskforge/CONTRACTS.md` this session; resolve
`$SCRIPT`; guard on readiness

via SKILL·HOT · 数据来自 GitHub 公开信息 · 原文版权归作者所有