Build outline

Skill eugenelim/agent-ready-repo/packs/desk-research/.apm/skills/build-outline

The complete AI operating model for software teams — from first idea to production. Three peer-supervised loops (discovery → build → release) over a catalogue of curated packs: skills, subagents, and hooks, each installed in one line. It's npm for your coding agent. Any agent, any stack — Claude Code, Codex, Cursor, Copilot, Gemini, Kiro.

Install

npx -y skills add eugenelim/agent-ready-repo --skill build-outline

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

14 stars14 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Decompose a research question into the sub-questions a thorough answer must address. Builds the outline that `/source-map` then populates and `/desk-research` then synthesises against. Grounded in STORM's outline stage (multi-perspective topic decomposition) and PRISMA's PICO framework (Population, Intervention, Comparison, Outcome — the systematic-review decomposition). Produces `<topic-slug>-outline.md` listing each sub-question with a brief rationale. Depth cues — `quickly`, `top three`, `briefly`, `summary only` for the must-answer few; `comprehensively`, `exhaustively`, `in depth`, `extensive` to chase second-order sub-questions.

SKILL.md

3.6 KB, as published. Nobody here has run it

/build-outline

The pre-research scaffold. Decomposes a question into sub-questions so the synthesis step has a structure to fill.

When to invoke

Before standard or deep /desk-research on a broad question.
As the first step in the survey pipeline (/build-outline → /source-map → /desk-research).
Not for narrow factual questions — those go straight to /desk-research.

Methodology

Two convergent disciplines:

STORM outline stage — STORM (Stanford's Synthesis of Topic Outlines through Retrieval and Multi-perspective question-asking) builds Wikipedia-style outlines by surveying adjacent topics, then asking what sections such an article would need. The pack borrows the survey-then-decompose shape: decomposition follows from what adjacent material covers, not from what the model assumes.
PRISMA PICO — the systematic-review framework decomposes a clinical question along four axes: Population, Intervention, Comparison, Outcome. PICO generalises beyond medicine: every research question has a target, a variable, an alternative, and a criterion. The pack borrows the axis-decomposition shape.

Procedure

Restate the question. Identify the axes that matter — PICO is the medical case; for a software question the axes might be "system / change / alternative / failure mode".
Survey adjacent material (STORM step) — what do good answers to adjacent questions look like? What sections do they have?
Enumerate sub-questions — each one is a question the final <topic-slug>-survey.md must answer. Tag each with a one-sentence rationale.
Order the sub-questions — context-first, then comparisons, then trade-offs, then conclusions.
Write <topic-slug>-outline.md — <topic-slug> is the kebab-case topic slug; the naming rule lives in the /desk-research skill body (§ Typed, topic-named artifacts).

`<topic-slug>-outline.md` output schema

# Outline — <main question>

## Sub-question 1: <question>

**Rationale:** <one sentence on why this matters to the main question>.

## Sub-question 2: <question>

(same shape)

## Open / second-order sub-questions

- <question the main answer might raise but doesn't itself answer>.

Citation discipline

Sub-question rationales are arguments, not assertions — mark [synthesis] when they integrate across cited material, [inference] when they deduce from precedent. Citations attach when a sub-question is justified by a specific source (e.g., "PRISMA recommends this decomposition" → cite the PRISMA handbook).

Depth cues

quickly, top three, briefly, summary only — return the must-answer few sub-questions only; skip open / second-order.
comprehensively, exhaustively, in depth, extensive — chase second-order sub-questions; surface the questions a thorough answer raises but doesn't itself resolve.