Harness Long-Running Task Loop

What it does

When your agent uses it

• Running a long backlog of tasks autonomously over many wake-ups
• Waiting on CI between cycles with a short pacing interval
• Overnight batch runs with a long interval between wake-ups
• Checking on or stopping a running harness loop

Example prompts

• “Run /harness-loop all with at most 3 cycles.”
• “Loop through tasks 41.1 to 41.3 with CI pacing.”
• “Start the overnight batch for the roadmap plan using night pacing.”
• “Show the status of the running harness loop, then stop it.”

Requirements

• Claude Code with /loop and ScheduleWakeup
• The harness-work skill from the same plugin
• A plans/manifest.json file when using named plans
• Pre-approved tools (allowed-tools): Read, Edit, Bash, Task, ScheduleWakeup, mcp__harness__harness_mem_resume_pack, mcp__harness__harness_mem_record_checkpoint

Workflow steps

3 steps, taken from the first numbered list in SKILL.md.

1. 1CLAUDE_PLUGIN_ROOT があり、scripts/ を含むならそれを使う
2. 2CLAUDE_PLUGIN_ROOT がなければ、CLAUDE_SKILL_DIR から plugin bundle root を逆算する
3. 3どちらでも解けない場合は停止し、CLAUDE_PLUGIN_ROOT を plugin bundle root に設定して再実行する

What it can do on your machine

Read from SKILL.md and the folder at commit 2b2b748. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves these tools, so the agent can use them without asking each time:

  • Read
  • Edit
  • Bash
  • Task
  • ScheduleWakeup
  • mcp__harness__harness_mem_resume_pack
  • mcp__harness__harness_mem_record_checkpoint

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  Shell commands in SKILL.md call:

  • bash

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  No URLs in SKILL.md.

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names no API keys, tokens, secrets or passwords.

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Harness Long-Running Task Loop loads about 2.3k tokens when it runs, and up to ~9.5k if it reads all its reference files. Until then it costs about 77 tokens; SKILL.md has 488 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~77

When it runs · the whole SKILL.md, loaded when a task matches

~2.3k

With references · SKILL.md plus every file in references/, read only if the agent opens them

~9.5k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

• NotePre-approves every shell command (allowed-tools: Bash)SKILL.md

  allowed-tools: Read, Edit, Bash, Task, ScheduleWakeup, mcp__harness__harness_mem_resume_pack, mcp__harness__harness

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from Chachamaru127/claude-code-harness at commit 2b2b748, republished under its MIT licence (© Chachamaru127). 488 words, ~2,321 tokens.

name

harness-loop

description

Long-running task loop using /loop (Claude Code dynamic mode) and ScheduleWakeup to re-enter with fresh context on each wake-up. Internally invokes harness-work through Agent. Trigger: long-running, loop, wake-up, autonomous. Do NOT load for: one-shot task execution, review, release, planning.

allowed-tools

Read, Edit, Bash, Task, ScheduleWakeup, mcp__harness__harness_mem_resume_pack, mcp__harness__harness_mem_record_checkpoint

description-en

Long-running task loop using /loop (Claude Code dynamic mode) and ScheduleWakeup to re-enter with fresh context on each wake-up. Internally invokes…

description-ja

長時間タスクを /loop と ScheduleWakeup で wake-up 毎に fresh context で再入実行。harness-work を内部で Agent 呼び出し。長時間、ループ、loop、wake-up、autonomous に対応。

kind

workflow

purpose

Re-enter long-running Plans.md execution with fresh context

trigger

long-running, loop, wake-up, autonomous

shape

delegate

role

orchestrator

base

harness-work

pair

harness-sync

owner

harness-core

wake-up
  │
  ▼
[Step 0] plugin bundle root を HARNESS_PLUGIN_ROOT に解決
  CLAUDE_PLUGIN_ROOT が有効ならそれを使用
  なければ CLAUDE_SKILL_DIR から plugin bundle root を逆算
  ※ host project cwd の scripts/ は参照しない
  │
  ▼
[Step 1] Plans.md を先に読む
  cc:WIP / cc:TODO の先頭タスクを特定（task_id を得る）
  ※ 未完了タスクなし → ループ終了（正常完了）
  │
  ▼
[Step 2] sprint-contract 存在確認 & 生成
  .claude/state/contracts/${task_id}.sprint-contract.json の有無を確認
  無ければ node "${HARNESS_PLUGIN_ROOT}/scripts/generate-sprint-contract.js" ${task_id} で生成
  生成直後（初回のみ）: bash "${HARNESS_PLUGIN_ROOT}/scripts/enrich-sprint-contract.sh" <contract-path> \
    --check "wake-up 自動承認（harness-loop のため DoD を reviewer 観点で確認）" \
    --approve  ← draft → approved に昇格
  （既存 contract は approved 済みのためスキップ）
  │
  ▼
[Step 3] contract readiness チェック
  bash "${HARNESS_PLUGIN_ROOT}/scripts/ensure-sprint-contract-ready.sh" <contract-path>
  │
  ▼
[Step 4] Resume pack 再読込
  harness-mem resume-pack（コンテキスト再注入）
  │
  ▼
[Step 4.5] Advisor consult（必要時のみ）
  高リスク task の初回実行前 / 同じ原因の 2 回目失敗後 / plateau 直前で
  `advisor-request.v1` を組み立てて相談する
  │
  ├── PLAN        → 次回 executor prompt に advice を prepend
  ├── CORRECTION  → 局所修正の指示として再実行
  └── STOP        → その場で loop を停止し、理由を残す
  │
  ▼
[Step 5] 1 タスクサイクル実行
  worker_result = Agent(
      subagent_type="claude-code-harness:worker",  # worker エージェント（harness-work ではない）
      prompt="タスク: ${task_id}\n内容と目的: <選択済み Plans.md の内容・理由>\n担当範囲: <owned paths と除外範囲>\nDoD: <Plans.md から抽出>\nplan/spec: <選択済みパス>\ncontract_path: ${CONTRACT_PATH}\n証拠と前回結果: <実行結果・失敗履歴・advisor response>\n原依頼と承認の参照: <適用範囲と参照先>\nmode: breezing\n担当範囲で方法を選び、他担当の変更を戻さず完了してください。判断理由、検証コマンドの実結果と証拠、未確認点を返してください。",
      isolation="worktree",
      run_in_background=false
  )
  # worker_result: { commit, branch, worktreePath, files_changed, summary }
  │
  ▼
[Step 5.5] Lead レビュー実行
  diff_text = git show worker_result.commit
  verdict = codex_exec_review(diff_text) or reviewer_agent_review(diff_text)
  ※ 詳細は flow.md 参照
  │
  ▼
[Step 5.6] APPROVE → main に cherry-pick / REQUEST_CHANGES → 修正ループ（contract の max_iterations 回、デフォルト 3）
  APPROVE: git cherry-pick → Plans.md を cc:完了 [{hash}] に更新 → feature branch 削除
  REQUEST_CHANGES x MAX_REVIEWS 後も否決: エスカレーション
  ※ 詳細は flow.md 参照
  │
  ▼
[Step 6] plateau 判定
  bash "${HARNESS_PLUGIN_ROOT}/scripts/detect-review-plateau.sh" ${current_task_id}
  │
  ├── PIVOT_REQUIRED（exit 2）  → ループ停止 + ユーザーエスカレーション
  ├── INSUFFICIENT_DATA（exit 1）→ 続行
  └── PIVOT_NOT_REQUIRED（exit 0）→ 続行
  │
  ▼
[Step 7] サイクル数チェック
  │
  ├── cycles >= max_cycles → ループ停止（上限到達）
  │
  ▼
[Step 8] checkpoint 記録
  harness_mem_record_checkpoint(
      session_id, title, content=サイクル結果サマリ
  )
  │
  ▼
[Step 9] 次 wake-up 予約
  ScheduleWakeup(
      delaySeconds=<pacing値>,
      prompt="/harness-loop <同じ引数>",
      reason="サイクル {N}/{max} 完了 — 次タスクへ"
  )

{
  "session_id": "<セッション ID>",
  "title": "harness-loop cycle {N}/{max}: {タスク名}",
  "content": "cycle_result の 1 行サマリ + commit hash"
}

© Chachamaru127, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 1 other file (references) in skills/harness-loop of Chachamaru127/claude-code-harness.

• SKILL.md
• references/flow.md

Open the folder on GitHubat commit 2b2b748