本文へ移動
cccskills
無料GitHub で公開

ralph-specum-tasks

This skill should be used only when the user explicitly asks to use `$ralph-specum-tasks`, or explicitly asks Ralph Specum in Codex to run the tasks phase.

インストール方法を見る

含まれるファイル(2)

  • SKILL.md6.3 KB
  • agents/openai.yaml1.1 KB

SKILL.md(原文)

インストールする前に、エージェントに与えられる指示の中身を確認できます。

Ralph Specum Tasks

You are a coordinator, not a task planner -- delegate ALL work to a task-planner sub-agent.

Derive RALPH_CODEX_PLUGIN_ROOT from this loaded skill by resolving two parent directories from the SKILL.md directory. Never derive it from the project working directory.

Contract

  • Resolve the active spec by explicit path, exact name, or .current-spec
  • Require requirements.md and design.md
  • Merge state fields only
  • Keep the Ralph disk contract unchanged

Action

  1. Resolve the active spec. If none exists, stop.
  2. Require requirements.md and design.md. Read research.md when present, .progress.md, and current state.
  3. Run phase_gate.py mode through "$RALPH_CODEX_PLUGIN_ROOT/scripts/phase_gate.py" with STATE and exact --quick, exact --interactive, or no flag. Reject both, -q, variants, and natural-language substitutes.
  4. In interactive mode, require artifact approval for the current design.md before starting tasks. Exact quick mode continues with the validated artifact.
  5. Run prototype record selection with the resolved basePath before generation:
    python3 "$RALPH_CODEX_PLUGIN_ROOT/scripts/prototype_records.py" select-downstream --base-path "$BASE_PATH" --state "$BASE_PATH/.ralph-state.json"
    
  6. Include only affected, valid, gateApproved: true, non-superseded records returned by the selector. Exclude malformed, superseded, skipped, failed, inconclusive, cancelled, and normal-mode excluded records.
  7. Stop before generation when selection reports an activePrototypes blocker for tasks. Name the active ID and route resume through $ralph-specum-prototype. Allow proven unrelated prototypes when the selector reports no task dependency.
  8. Stop when selection reports stale design.md, stale task indexes, or a stale upstream artifact that design depends on. Route to the earliest stale phase and do not plan from stale design.
  9. Clear any prior approval gate by merging awaitingApproval: false before generation.
  10. Respect granularity from state. Allow --tasks-size fine|coarse to override it. Treat task sizing as administration, not an interview question. In exact quick mode, default unset granularity to fine.
  11. When research.md exists, require skill discovery pass 2 against the goal plus final research. When it is absent, require pass 1 against the goal alone. Run the applicable pass when the state lacks its revision. Select explicitly named skills and record harness-shadowed duplicates.
  12. Load "$RALPH_CODEX_PLUGIN_ROOT/skills/interview-framework-codex/SKILL.md", its required algorithm and domain-modeling references, and all selected domain contracts in both interactive and quick mode. In interactive mode, follow the algorithm for critical delivery slicing, dependency order, rollout risk, and verification thresholds. Inspect commands, file layout, and existing test tools instead of asking.
  13. In interactive mode, require explicit approve and delegate; in exact quick mode, record bypassed_quick. In both modes, run phase_gate.py check-delegation with the current loaded-manifest identity before creating the child. Apply the shared hard-transition invariant before task-planner dispatch. A failed check-delegation in either mode ends this invocation before phase transition, child dispatch, or target-artifact write; after a normal-mode failure, the next explicit invocation creates a fresh manifest/interview identity and does not create an alternate quick bypass. After an exact --quick delegation failure, the next explicit invocation reruns discovery, records a fresh phaseSkillLoad and interview identity, and does not reuse a terminal bypassed_quick interview or its discovery revision. Only a matching in-progress collecting or awaiting_confirmation interview is resumable.
  14. Delegate task planning to a task-planner sub-agent. Pass the absolute helper path, state path, identity tuple, unique teammate dispatch identity, verbatim manifest, requirements, design, research, selected prototype evidence, the clean blocker/stale-gate result, and interview context. The child reloads and records the manifest, passes check-agent-write with that unique identity, and writes tasks.md. Do NOT write tasks.md yourself.
  15. Read the sub-agent's output and validate it exists.
  16. Count tasks and merge state with:
  • phase: "tasks"
  • awaitingApproval: true (or false when --quick is active)
  • taskIndex: first incomplete or totalTasks
  • totalTasks: counted tasks
  1. Update .progress.md with the phase breakdown, next milestone, blockers, next step, chosen granularity, skill discovery, and verification strategy.
  2. If spec commits are enabled, commit only the spec artifacts.

Stop Behavior

  • Without --quick: STOP HERE. Display the walkthrough summary and approval prompt. Do NOT continue to implementation. Wait for the user to explicitly approve and request the next phase.
  • With exact --quick: Record the quick bypass, review, then continue directly into implementation.

Output Shape

Use atomic tasks with exact file targets, explicit success criteria, verification commands, and commit messages. Preserve POC-first ordering. Support [P] markers for safe parallel work, [VERIFY] checkpoints, and VE tasks when end-to-end verification is part of the plan.

Response Handoff

  • After writing tasks.md, name tasks.md and summarize the task plan briefly.
  • When normalized quickMode is false, end with exactly one explicit choice prompt:
    • approve current artifact
    • request changes
    • continue to implementation
  • Treat continue to implementation as approval of tasks.md.
  • With exact --quick, do not show this prompt; continue directly to implementation after the gates succeed.
  • During artifact review, apply the changes immediately delegates already-recorded feedback through a new unique dispatch, redisplays the artifact, and stays at this approval gate. Ask one focused change question only when no feedback is pending. Control-only continue, proceed, and go ahead approve nothing.

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

同じリポジトリのスキル

概要と使いどころ

This skill should be used when the user asks to "create a slash command", "add a command", "write a custom command", "define command arguments", "use command frontmatter", "organize commands", "create command with file references", "interactive command", "use AskUserQuestion in command", or needs guidance on slash command structure, YAML frontmatter fields, dynamic arguments, bash execution in commands, user interaction patterns, or command development best practices for Claude Code.

日本語の概要は準備中です。原文の説明を表示しています。

tzachbon/smart-ralph5592026年9月16日 更新

This skill should be used when generating spec artifacts (research.md, requirements.md, design.md, tasks.md), formatting agent output, structuring phase results, or when any Ralph agent needs guidance on concise, scannable output formatting. Applies to all Ralph spec phase agents.

日本語の概要は準備中です。原文の説明を表示しています。

tzachbon/smart-ralph5592026年9月16日 更新

Output rules for all agents - concise, scannable, actionable. Based on Matt Pocock's planning principles.

日本語の概要は準備中です。原文の説明を表示しています。

tzachbon/smart-ralph5592026年9月16日 更新

Core principle that the main agent is a coordinator, not an implementer. All work must be delegated to subagents.

日本語の概要は準備中です。原文の説明を表示しています。

tzachbon/smart-ralph5592026年9月16日 更新

This skill should be used when the user asks to "create a hook", "add a PreToolUse/PostToolUse/Stop hook", "validate tool use", "implement prompt-based hooks", "use ${CLAUDE_PLUGIN_ROOT}", "set up event-driven automation", "block dangerous commands", or mentions hook events (PreToolUse, PostToolUse, Stop, SubagentStop, SessionStart, SessionEnd, UserPromptSubmit, PreCompact, Notification). Provides comprehensive guidance for creating and implementing Claude Code plugin hooks with focus on advanced prompt-based hooks API.

日本語の概要は準備中です。原文の説明を表示しています。

tzachbon/smart-ralph5592026年9月16日 更新

This skill should be used when a Ralph phase must identify critical user decisions, run a layered grill, persist partial answers, obtain explicit approval, or resume an interrupted phase interview before delegating artifact work.

日本語の概要は準備中です。原文の説明を表示しています。

tzachbon/smart-ralph5592026年9月16日 更新

tzachbon のスキルをすべて見る

このスキルの問題を報告する