跳至正文
OpenAgentSkill
浏览已开放 · 目录记录日期 2026-11-21 · 自动更新和账户服务暂时关闭。
搜索与研究

rootnode-output-blocks

Tested output format specifications for Claude prompts. Use when the user wants a specific deliverable structure — retrieving, reviewing, customizing, or building output formatting. Trigger on: "give me the Executive Brief format," "show me the output template for," "I need the Decision Matrix structure," "show me available output formats," "build a custom output format for." Provides 10 ready-to-use structures (executive briefs, technical designs, research summaries, implementation plans, decision matrices, competitive analyses, post-mortems, stakeholder updates, strategic memos, process documentation) plus a custom template, with per-section length guidance and format constraints tested against Claude formatting defaults. If unsure which format fits, use rootnode-block-selection first if available. Do NOT use for evaluating existing prompts (use rootnode-prompt-validation if available) or choosing reasoning/identity approaches (use rootnode-block-selection if available).

rootnode · 已有静态检查记录

安装与使用

运行命令前先阅读来源与依赖说明。复制命令不会自动安装或执行。

npx skills add drayline/rootnode-skills --skill rootnode-output-blocks

技能说明


name: rootnode-output-blocks description: >- Tested output format specifications for Claude prompts. Use when the user wants a specific deliverable structure — retrieving, reviewing, customizing, or building output formatting. Trigger on: "give me the Executive Brief format," "show me the output template for," "I need the Decision Matrix structure," "show me available output formats," "build a custom output format for." Provides 10 ready-to-use structures (executive briefs, technical designs, research summaries, implementation plans, decision matrices, competitive analyses, post-mortems, stakeholder updates, strategic memos, process documentation) plus a custom template, with per-section length guidance and format constraints tested against Claude formatting defaults. If unsure which format fits, use rootnode-block-selection first if available. Do NOT use for evaluating existing prompts (use rootnode-prompt-validation if available) or choosing reasoning/identity approaches (use rootnode-block-selection if available). license: Apache-2.0 metadata: author: rootnode version: "4.0.0" original-source: "BLOCK_LIBRARY_OUTPUT.md"

Output Format Selection for Claude Prompts

Calibration: Tier 1 (Model-compatible) - runs cleanly on the current dual-primary tier (Opus 5, Sonnet 5) as well as Haiku 4.5 with extended thinking. Structured retrieval, rule evaluation, or template lookup - output shape does not depend on model class. Correct-shape output also on Opus 4.8 (fallback-graceful) and Sonnet 4.6 (legacy-graceful). See repository README for model compatibility.

Tested output format specifications that control what Claude's deliverable looks like — structure, sections, length, and format. Without an explicit output format, Claude defaults to its own formatting preferences, which often means overlong prose with excessive bullet points.

When to Use This Skill

Use this Skill when you need to:

  • Choose an output format for a Claude prompt you are building
  • Structure a deliverable (brief, memo, plan, analysis, etc.)
  • Design a custom output format for a deliverable type not covered here
  • Fix output quality problems caused by missing or vague format instructions

This Skill provides the format specification (Layer 5 of a well-structured prompt). It does not cover identity, reasoning methodology, or quality control — for those, see rootnode-block-selection or rootnode-prompt-compilation if available.

Quick-Reference: Deliverable Type → Format

You need...Use this formatReference file
A decision or recommendation for senior leadershipExecutive Briefreferences/executive-formats.md
A strategic argument or policy position for decision-makersStrategic Memoreferences/executive-formats.md
A progress report or status update for stakeholdersStakeholder Updatereferences/executive-formats.md
An architecture proposal or RFC for technical reviewersTechnical Design Documentreferences/technical-formats.md
A workflow, runbook, or SOP for process executorsProcess Documentationreferences/technical-formats.md
Evidence synthesis or research findingsResearch Summaryreferences/analytical-formats.md
A structured comparison of options against criteriaDecision Matrixreferences/analytical-formats.md
A market positioning or competitor assessmentCompetitive Analysisreferences/analytical-formats.md
A project plan with phases, dependencies, and timelinesImplementation Planreferences/operational-formats.md
An incident analysis or project retrospectivePost-Mortem / Retrospectivereferences/operational-formats.md
Something not listed aboveBuild a custom format (see below)—

Selection principle: Match the format to the audience and their primary question. Executives ask "what should we do?" → Executive Brief or Strategic Memo. Engineers ask "how does this work?" → Technical Design Document. Operators ask "what do I do next?" → Process Documentation or Implementation Plan. Analysts ask "what did we find?" → Research Summary.

How to Use a Format Specification

Each format in the reference files includes:

  1. Use-when guidance — the scenario and audience it fits
  2. Complete XML specification — copy directly into your prompt's output layer
  3. Per-section length guidance — word counts, sentence counts, or item counts for each section
  4. Watch-for notes — Claude's common failure modes with that format and specific countermeasure language to add

To apply a format: copy the <output_format> XML block into your prompt. Customize section names and lengths to your situation. Remove sections that are not relevant. The specifications are tested starting points — adjust to match what your audience actually needs.

Critical: Controlling Claude's Formatting Defaults

Claude has strong formatting tendencies that output specifications must actively manage:

Bullet point overuse. Claude defaults to bullet points for almost everything. If you want prose, say "write in prose" explicitly in the relevant section. If you want a table, say "present as a table with columns: X, Y, Z."

Length inflation. Claude tends to write longer than necessary. Per-section length guidance (e.g., "2-3 sentences" or "1 paragraph") is more effective than total word counts alone. Specify both when possible.

Burying the lead. Claude often builds up to its conclusion rather than leading with it. Executive-facing formats (Executive Brief, Strategic Memo, Stakeholder Update) all specify lead-with-the-answer structure — enforce this explicitly if Claude reverts.

Hedging on assessments. In formats requiring judgment (Competitive Analysis, Decision Matrix, Post-Mortem), Claude may soften findings. The watch-for notes in each format specification include countermeasure language to add when this happens.

Building Custom Output Formats

When no exist

阅读完整来源 ↗