菜单
Run high-signal autonomous coding loops with Soulforge (feature-dev/bugfix/review-loop) using strict worktree isolation, review gates, and scoped fix cycles.
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/javimosch/open-claw-skills --skill soulforge命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
基于 SOC 职业分类
正在显示 SKILL.md
| name | soulforge |
| description | Run high-signal autonomous coding loops with Soulforge (feature-dev/bugfix/review-loop) using strict worktree isolation, review gates, and scoped fix cycles. |
| metadata | {"openclaw":{"emoji":"🔥","requires":{"bins":"[Truncated]","env":"[Truncated]"}}} |
This is not a full engine reference. This is the operating playbook for getting high-quality autonomous coding outcomes with Soulforge.
Use Soulforge when you want: plan → implement → verify/test → PR → review/fix loops with minimal babysitting.
Preferred workflows:
feature-dev for end-to-end feature deliverybugfix for diagnose-first, surgical fixesreview-loop for tightening an existing PR until clean<repo>/worktrees/SEPARATE.--workdir is omitted, Soulforge can auto-provision a worktree under <repo>/worktrees/....approve/reject is gone.soulforge complete ....type: pause.--callback-exec.{{run_id}}, {{step_id}}, {{step_status}}, {{status}}, {{task}}{{callback_message}} (step-level, preferred){{prompt}} remains for backward compatibility in pause scenariossoulforge run feature-dev "Implement <issue-url>.
Constraints: max 2 stories. DO NOT refactor unrelated modules." \
--workdir /abs/path/to/repo/worktrees/feat-xyz \
--callback-exec 'openclaw agent --session-key "agent:cpto:slack:channel:c0af7b05h28" --message "Soulforge {{run_id}} {{step_id}} {{step_status}}" --deliver'
soulforge run bugfix "Fix <issue-url> with failing test first; minimal patch only." \
--workdir /abs/path/to/repo/worktrees/fix-xyz \
--callback-exec 'openclaw agent --session-key "agent:cpto:slack:channel:c0af7b05h28" --message "Soulforge {{run_id}} {{step_id}} {{step_status}}" --deliver'
soulforge run review-loop "Review PR #123 and fix only in-scope findings." \
--workdir /abs/path/to/repo/worktrees/pr-123 \
--var pr_number=123 \
--callback-exec 'openclaw agent --session-key "agent:cpto:slack:channel:c0af7b05h28" --message "Soulforge {{run_id}} {{step_id}} {{step_status}}" --deliver'
Include:
If a PR loops repeatedly:
.soulforge-progress.md in worktree with exact outstanding fixesreview-loop constrained to remaining findingsAt review gate:
Long fix steps are normal for real refactors. Your job is quality control at gates, not interrupting active runs.
When code-review returns findings:
max_loops, spawn a fresh constrained review-loop run with a scope lock file.