Outline Skill: From Vague Idea to Structured Output

Where this fits

You're building an agent that helps users draft newsletters or sales pages. The user says "I want to write about remote work productivity" — and your agent produces a generic five-section outline that could apply to any topic. The problem isn't lack of writing ability; it's lack of structure. The outline skill solves this by running a structured interrogation that extracts the user's angle, audience, and format constraints before generating an outline. It's perfect for agents that need to produce format-specific skeletons for newsletters, posts, scenes, or scripts.

Why agents benefit

• Forces specificity: Instead of guessing, the agent asks targeted questions (e.g., "What's the primary outcome for the reader?") until the vague idea becomes a concrete brief.
• Respects format constraints: The skill knows the difference between a newsletter and a script — it adapts the interrogation and outline structure accordingly.
• Enforces anti-slop rules: It reads dependency files that forbid fabrication and over-correction, so the outline is grounded in the user's actual input, not hallucinated patterns.
• Reusable across domains: Whether it's a sales page or a YouTube script, the same interrogation engine works, reducing the need to build separate prompt chains.
• Produces actionable output: The final outline is a tight, sectioned plan that the agent can pass directly to a writing step, eliminating back-and-forth.

Practical scenarios

Newsletter drafting assistant

A user subscribes to your agent to write a weekly newsletter. They say "I want to talk about AI tools." The agent uses outline to ask about the newsletter's typical structure, the target audience's pain points, and the desired call-to-action. The result is a three-part outline with specific bullet points that match the user's past issues.

Sales page for a new product

A startup founder needs a landing page but only has a feature list. The skill interrogates for the core promise, the primary objection, and the proof points. It then produces an outline with sections like "Problem," "Solution," "Social Proof," and "Risk Reversal" — tailored to the product's specifics.

Script scene for a video

A content creator wants to produce a 2-minute explainer video but has no narrative structure. The agent uses outline to extract the key message, the emotional hook, and the logical progression. The skill outputs a scene-by-scene breakdown with timing suggestions, making the subsequent scriptwriting step trivial.

Add it to your agent workflow

Install the skill directly from the EveryDay-Writer repository:

npx skills add Deupaxx/EveryDay-Writer --skill outline

Once installed, your agent can invoke it via /ew:outline or through the /ew dispatch. The skill reads its dependency chain (anti-AI rules, voice profile, etc.) automatically. Example usage in a prompt:

User: "I want to write a blog post about remote team communication."
Agent: (Runs outline skill) "Let's refine that. What's the primary audience? What specific pain point should the post address?"

The skill returns a structured outline that you can feed into your writing pipeline.

Compare before adopting

Before you commit, compare outline against other structuring methods:

• Quality signals: Does it actually produce non-generic outlines? Test with a few vague prompts and check if the questions are context-aware.
• Maintenance freshness: The repo has 28 stars and is actively updated (version 0.3.0). Check the commit history and issue responses.
• Alternatives: There are other prompting techniques like chain-of-thought or few-shot templates. outline offers a dedicated interrogation flow, but it may be overkill for simple tasks.
• Workflow fit: Does it integrate with your existing agent's toolchain? The dependency on voice-profile and anti-slop files might require adjustment if your agent has its own style rules.

Why it is worth tracking

The outline skill stands out because it addresses a fundamental flaw in many writing agents: they jump to prose without a solid skeleton. Its structured interrogation approach is a practical pattern that can be adapted to other content types. With 28 stars and a MIT license, it's a community-driven project worth monitoring — especially if you're building agents that need to produce long-form content. As the EveryDay-Writer suite grows, expect more skills that follow the same rigorous dependency chain, making it a valuable reference for agent design. For a deeper look, check the skill page on OpenAgentSkill: https://www.openagentskill.com/skills/deupaxx-outline