ワンクリックで
forge-execute
Use when you have a written implementation plan to execute in a separate session with review checkpoints
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
メニュー
Use when you have a written implementation plan to execute in a separate session with review checkpoints
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Use when executing plans through specialized subagents with evidence-gated review loops
Use when work needs an explicit delivery contract, iterative repair, bounded autonomy, or resume-safe state
Use when a user asks to continue, resume, pick up prior loop work, or inspect previous loop state
Use when work needs codebase orientation, impact analysis, existing-pattern discovery, or risk scouting before planning or implementation
Use when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always
Use when implementation direction depends on unresolved product/design choices, multiple viable approaches, or owner approval on a design.
| name | forge:execute |
| hidden | true |
| description | Use when you have a written implementation plan to execute in a separate session with review checkpoints |
Load plan, review critically, execute all tasks, report when complete.
Announce at start: "I'm using the forge:execute skill to implement this plan."
Note: Tell your human partner that work of this complexity works much better with subagent access. The quality will be significantly higher if run on a platform with subagent support (such as Claude Code or Codex). If subagents are available, use forge:subagent instead of this skill.
task tool and proceedFor each task:
- [ ] checkbox to - [x] and
append a line to the plan's ## Execution Progress section (see forge:plan's
"Living Plan Document" contract). The line MUST include: commit SHA, verify
verdict, ISO-8601 timestamp. If verify failed and you are iterating, leave the
checkbox - [ ] and append an iteration-N line to ## Loop Trace instead.Skipping the writeback is a false completion. The task is not "done" until the plan document reflects it.
After all tasks complete and verified:
skill("forge:report") to write the final report (summarizes what was built in human-readable form)skill("forge:merge") on completionSTOP executing immediately when:
Use forge:ask to present the blocker and options rather than describing it in free text. If no user is available, resolve the blocker with your best judgment and continue.
Return to Review (Step 1) when:
Don't force through blockers - stop and ask.
Required workflow skills: