# Idea Discovery

> Workflow 1: Full idea discovery pipeline to go from a broad research direction to validated, pilot-tested ideas. Use when user says \"找idea全流程\", \"idea discovery pipeline\", \"从零开始找方向\", or wants the complete idea exploration workflow.

- **Type:** Skill
- **Install:** `agentstack add skill-wanshuiyin-auto-claude-code-research-in-sleep-idea-discovery`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [wanshuiyin](https://agentstack.voostack.com/s/wanshuiyin)
- **Installs:** 0
- **Category:** [AI & ML](https://agentstack.voostack.com/c/ai-and-ml)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [wanshuiyin](https://github.com/wanshuiyin)
- **Source:** https://github.com/wanshuiyin/Auto-claude-code-research-in-sleep/tree/main/skills/idea-discovery

## Install

```sh
agentstack add skill-wanshuiyin-auto-claude-code-research-in-sleep-idea-discovery
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# Workflow 1: Idea Discovery Pipeline

Orchestrate a complete idea discovery workflow for: **$ARGUMENTS**

## Overview

This skill chains sub-skills into a single automated pipeline:

```
/research-lit → /idea-creator → /novelty-check → /research-review → /research-refine-pipeline
  (survey)      (brainstorm)    (verify novel)    (critical feedback)  (refine method + plan experiments)
```

Each phase builds on the previous one's output. The final deliverables are a validated `idea-stage/IDEA_REPORT.md` with ranked ideas, plus a refined proposal (`refine-logs/FINAL_PROPOSAL.md`) and experiment plan (`refine-logs/EXPERIMENT_PLAN.md`) for the top idea.

## Constants

- **PILOT_MAX_HOURS = 2** — Skip any pilot experiment estimated to take > 2 hours per GPU. Flag as "needs manual pilot" in the report.
- **PILOT_TIMEOUT_HOURS = 3** — Hard timeout: kill any running pilot that exceeds 3 hours. Collect partial results if available.
- **MAX_PILOT_IDEAS = 3** — Run pilots for at most 3 top ideas in parallel. Additional ideas are validated on paper only.
- **MAX_TOTAL_GPU_HOURS = 8** — Total GPU budget across all pilots. If exceeded, skip remaining pilots and note in report.
- **AUTO_PROCEED = true** — If user doesn't respond at a checkpoint, automatically proceed with the best option after presenting results. Set to `false` to always wait for explicit user confirmation.
- **REVIEWER_MODEL = `gpt-5.5`** — Model used via Codex MCP. Must be an OpenAI model (e.g., `gpt-5.5`, `o3`, `gpt-4o`). Passed to sub-skills.
- **OUTPUT_DIR = `idea-stage/`** — All idea-stage outputs go here. Create the directory if it doesn't exist.
- **ARXIV_DOWNLOAD = false** — When `true`, `/research-lit` downloads the top relevant arXiv PDFs during Phase 1. When `false` (default), only fetches metadata. Passed through to `/research-lit`.
- **COMPACT = false** — When `true`, generate compact summary files for short-context models and session recovery. Writes `idea-stage/IDEA_CANDIDATES.md` (top 3-5 ideas only) at the end of this workflow. Downstream skills read this instead of the full `idea-stage/IDEA_REPORT.md`.
- **RENDER_HTML = true** — When `true` (default), auto-render `idea-stage/IDEA_REPORT.md` to HTML at workflow end via `/render-html`. Uses `--no-review` (the source MD already went through novelty + cross-model review during Phase 3). Set `false` to skip, or pass `— render html: false`.
- **REF_PAPER = false** — Reference paper to base ideas on. Accepts: local PDF path, arXiv URL, or any paper URL. When set, the paper is summarized first (`idea-stage/REF_PAPER_SUMMARY.md`), then idea generation uses it as context. Combine with `base repo` for "improve this paper with this codebase" workflows.

> 💡 These are defaults. Override by telling the skill, e.g., `/idea-discovery "topic" — ref paper: https://arxiv.org/abs/2406.04329` or `/idea-discovery "topic" — compact: true`.

## Pipeline

### Phase 0: Load Research Brief (if available)

Before starting any other phase, check for a detailed research brief in the project:

1. Look for `RESEARCH_BRIEF.md` in the project root (or path passed as `$ARGUMENTS`)
2. If found, read it and extract:
   - Problem statement and context
   - Constraints (compute, data, timeline, venue)
   - What the user already tried / what didn't work
   - Domain knowledge and non-goals
   - Existing results (if any)
3. Use this as the primary context for all subsequent phases — it replaces the one-line prompt
4. If both `RESEARCH_BRIEF.md` and a one-line `$ARGUMENTS` exist, merge them (brief takes priority for details, argument sets the direction)

If no brief exists, proceed normally with `$ARGUMENTS` as the research direction.

> 💡 Create a brief from the template: `cp templates/RESEARCH_BRIEF_TEMPLATE.md RESEARCH_BRIEF.md` — keep it to ~1-2 pages (4-8k chars); long material goes in separate files referenced by path.

### Phase 0.5: Reference Paper Summary (when REF_PAPER is set)

**Skip entirely if `REF_PAPER` is `false`.**

Summarize the reference paper before searching the literature:

1. **If arXiv URL** (e.g., `https://arxiv.org/abs/2406.04329`):
   - Invoke `/arxiv "ARXIV_ID" — download` to fetch the PDF
   - Read the first 5 pages (title, abstract, intro, method overview)

2. **If local PDF path** (e.g., `papers/reference.pdf`):
   - Read the PDF directly (first 5 pages)

3. **If other URL**:
   - Fetch and extract content via WebFetch

4. **Generate `idea-stage/REF_PAPER_SUMMARY.md`**:

```markdown
# Reference Paper Summary

**Title**: [paper title]
**Authors**: [authors]
**Venue**: [venue, year]

## What They Did
[2-3 sentences: core method and contribution]

## Key Results
[Main quantitative findings]

## Limitations & Open Questions
[What the paper didn't solve, acknowledged weaknesses, future work suggestions]

## Potential Improvement Directions
[Based on the limitations, what could be improved or extended?]

## Codebase
[If `base repo` is also set: link to the repo and note which parts correspond to the paper]
```

**🚦 Checkpoint:** Present the summary to the user:

```
📄 Reference paper summarized:
- Title: [title]
- Key limitation: [main gap]
- Improvement directions: [2-3 bullets]

Proceeding to literature survey with this as context.
```

Phase 1 and Phase 2 will use `idea-stage/REF_PAPER_SUMMARY.md` as additional context — `/research-lit` searches for related and competing work, `/idea-creator` generates ideas that build on or improve the reference paper.

### Phase 1: Literature Survey

Invoke `/research-lit` to map the research landscape. Idea discovery is exactly the place where Gemini's AI-driven broad coverage adds value, so include `gemini` as a source by default unless the user already specified an explicit `— sources:` directive in their idea-discovery invocation:

```
# If $ARGUMENTS already contains "— sources:", pass through unchanged
# (the user is in control of source selection):
/research-lit "$ARGUMENTS" — composed: idea-stage/IDEA_REPORT.md

# Otherwise (the common case), include gemini explicitly for broader discovery:
/research-lit "$ARGUMENTS" — sources: all, gemini — composed: idea-stage/IDEA_REPORT.md
```

`— composed: idea-stage/IDEA_REPORT.md` puts `/research-lit` in composed mode (see *Output hygiene* above): it returns the landscape for folding into the report instead of writing a standalone landscape file. The report doesn't exist yet at Phase 1 — the directive names the *forthcoming* canonical doc, and `/idea-creator` creates it in Phase 2.

If `gemini-cli` is not installed, `/research-lit` skips the Gemini source gracefully with a warning — no break to the pipeline. Users who want to force-disable Gemini in idea-discovery can pass `/idea-discovery "topic" — sources: all` explicitly (which becomes the literal source list, no auto-injection).

**What this does:**
- Search arXiv, Google Scholar, Semantic Scholar for recent papers
- Plus Gemini-driven broad discovery (sub-problem decomposition, naming variants, alias coverage) when `gemini-cli` is available
- Build a landscape map: sub-directions, approaches, open problems
- Identify structural gaps and recurring limitations
- Output a literature summary (saved to working notes)

**🚦 Checkpoint:** Present the landscape summary to the user. Ask:

```
📚 Literature survey complete. Here's what I found:
- [key findings, gaps, open problems]

Does this match your understanding? Should I adjust the scope before generating ideas?
(If no response, I'll proceed with the top-ranked direction.)
```

- **User approves** (or no response + AUTO_PROCEED=true) → proceed to Phase 2 with best direction.
- **User requests changes** (e.g., "focus more on X", "ignore Y", "too broad") → refine the search with updated queries, re-run `/research-lit` with adjusted scope, and present again. Repeat until the user is satisfied.

### Phase 2: Idea Generation + Filtering + Pilots

Invoke `/idea-creator` with the landscape context (and `idea-stage/REF_PAPER_SUMMARY.md` if available):

```
/idea-creator "$ARGUMENTS" — composed: idea-stage/IDEA_REPORT.md
```

`/idea-creator` owns `idea-stage/IDEA_REPORT.md` as the canonical deliverable; the `— composed:` directive tells it to fold the survey/novelty findings in rather than emitting `LIT_LANDSCAPE.md` / `RESEARCH_REVIEW.md` / `MANIFEST.md` alongside.

**What this does:**
- If `idea-stage/REF_PAPER_SUMMARY.md` exists, include it as context — ideas should build on, improve, or extend the reference paper
- Brainstorm 8-12 concrete ideas via GPT-5.5 xhigh
- Filter by feasibility, compute cost, quick novelty search
- Deep validate top ideas (full novelty check + devil's advocate)
- Run parallel pilot experiments on available GPUs (top 2-3 ideas)
- Rank by empirical signal
- Output `idea-stage/IDEA_REPORT.md`

**🚦 Checkpoint:** Present `idea-stage/IDEA_REPORT.md` ranked ideas to the user. Ask:

```
💡 Generated X ideas, filtered to Y, piloted Z. Top results:

1. [Idea 1] — Pilot: POSITIVE (+X%)
2. [Idea 2] — Pilot: WEAK POSITIVE (+Y%)
3. [Idea 3] — Pilot: NEGATIVE, eliminated

Which ideas should I validate further? Or should I regenerate with different constraints?
(If no response, I'll proceed with the top-ranked ideas.)
```

- **User picks ideas** (or no response + AUTO_PROCEED=true) → proceed to Phase 3 with top-ranked ideas.
- **User unhappy with all ideas** → collect feedback ("what's missing?", "what direction do you prefer?"), update the prompt with user's constraints, and re-run Phase 2 (idea generation). Before
  regenerating, read the already-tried directions (research-wiki Failed Ideas + any
  `.aris/runs/.iterations.jsonl`) and forbid a candidate too close to one already
  tried — enforced direction diversity; when an overnight heartbeat drives the run,
  record each chosen direction via `iteration_log.py note ... --direction ""`
  so later ticks can reject near-duplicates (see
  [`shared-references/external-cadence.md`](../shared-references/external-cadence.md) →
  Stall detection & forced structural pivot). Repeat until the user selects at least 1 idea.
- **User wants to adjust scope** → go back to Phase 1 with refined direction.

### Phase 3: Deep Novelty Verification

For each top idea (positive pilot signal), run a thorough novelty check:

```
/novelty-check "[top idea 1 description]"
/novelty-check "[top idea 2 description]"
```

**What this does:**
- Multi-source literature search (arXiv, Scholar, Semantic Scholar)
- Cross-verify with GPT-5.5 xhigh
- Check for concurrent work (last 3-6 months)
- Identify closest existing work and differentiation points

**Update `idea-stage/IDEA_REPORT.md`** with deep novelty results. Eliminate any idea that turns out to be already published.

### Phase 4: External Critical Review

For the surviving top idea(s), get brutal feedback:

```
/research-review "[top idea with hypothesis + pilot results]" — composed: idea-stage/IDEA_REPORT.md
```

In composed mode `/research-review` folds its conclusions into `idea-stage/IDEA_REPORT.md` and cites the `.aris/traces/…` path instead of writing a standalone review `.md` in the project root.

**What this does:**
- GPT-5.5 xhigh acts as a senior reviewer (NeurIPS/ICML level)
- Scores the idea, identifies weaknesses, suggests minimum viable improvements
- Provides concrete feedback on experimental design

**Update `idea-stage/IDEA_REPORT.md`** with reviewer feedback and revised plan.

### Phase 4.5: Method Refinement + Experiment Planning

After review, refine the top idea into a concrete proposal and plan experiments:

```
/research-refine-pipeline "[top idea description + pilot results + reviewer feedback]"
```

**What this does:**
- Freeze a **Problem Anchor** to prevent scope drift
- Iteratively refine the method via GPT-5.5 review (up to 5 rounds, until score ≥ 9)
- Generate a claim-driven experiment roadmap with ablations, budgets, and run order
- Output: `refine-logs/FINAL_PROPOSAL.md`, `refine-logs/EXPERIMENT_PLAN.md`, `refine-logs/EXPERIMENT_TRACKER.md`

**🚦 Checkpoint:** Present the refined proposal summary:

```
🔬 Method refined and experiment plan ready:
- Problem anchor: [anchored problem]
- Method thesis: [one sentence]
- Dominant contribution: [what's new]
- Must-run experiments: [N blocks]
- First 3 runs to launch: [list]

Proceed to implementation? Or adjust the proposal?
```

- **User approves** (or AUTO_PROCEED=true) → proceed to Final Report.
- **User requests changes** → pass feedback to `/research-refine` for another round.
- **Lite mode:** If reviewer score  Follow these shared protocols for all output files:
> - **[Output Composition Protocol](../shared-references/output-composition.md)** — ONE canonical deliverable per pipeline; fold sub-skill findings in, don't scatter overlapping `.md` files
> - **[Output Versioning Protocol](../shared-references/output-versioning.md)** — write timestamped file first, then copy to fixed name
> - **[Output Manifest Protocol](../shared-references/output-manifest.md)** — maintain `MANIFEST.md` only above the 15-artifact threshold (not "log every output")
> - **[Output Language Protocol](../shared-references/output-language.md)** — respect the project's language setting

### Output hygiene — ONE canonical doc, no duplicate MDs (REQUIRED)

This pipeline runs its sub-skills in **composed mode** (see
[`output-composition.md`](../shared-references/output-composition.md)): it owns a single
canonical deliverable and folds every sub-skill's findings into it rather than letting
each emit its own overlapping file. Concretely, for this workflow:

1. **`idea-stage/IDEA_REPORT.md` is the single canonical deliverable.** Sub-skills'
   intermediate findings (literature landscape, novelty notes, external review) are
   folded into it as sections/appendices — they do NOT become standalone files just
   because a sub-skill could emit one. If a sub-skill writes a scratch file, inline its
   unique content into the report and delete the scratch when the phase closes.
2. **Pass `— composed: idea-stage/IDEA_REPORT.md` to every sub-skill** (`/research-lit`,
   `/idea-creator`, `/research-review`) so they fold instead of scatter. This is the
   explicit signal; without it a sub-skill stays standalone by design.
3. **Refined-method outputs stay in `refine-logs/`** (`FINAL_PROPOSAL.md` /
   `EXPERIMENT_PLAN.md` / `EXPERIMENT_TRACKER.md`). Do NOT also restate them as separate
   files under `idea-stage/`; the report **links** to them, it does not copy them.
4. **No `MANIFEST.md`** for a handful of files — only above the 15-artifact threshold in
   [`output-manifest.md`](../shared-references/output-manifest.md).
5. **Pilot scratch is disposable:** keep the pilot script (reusable) + one results file
   (`pilot_results.jsonl` or a small summary). Delete launcher logs, smoke files, and
   redundant `*_summary.json` once the numbers are in the report.
6. **Cross-model review traces belong in `.aris/traces/…`** (the audit trail); do not
   ALSO keep a human-facing copy under `idea-stage/` — cite the trace path from the report.
7. **Before finishing,** the `idea-stage/` top level should be roughly: `IDEA_REPORT.md`
   (+ `.html`), the pilot script + results, and the `refine-logs/` dir. Nothing else
   unless it carries content not in the report.

## Render HTML view (auto, when `RENDER_HTML = true`)

After Phase 4 finalizes `idea-stage/IDEA_REPORT.md` (and the optional `IDEA_CANDIDATES.md`), invoke `/render-html` on the report so the user has a single-file HTML view for tablet / phone reading:

```
/render-html "idea-stage/IDEA_REPORT.md" --no-review
```

`--no-review` is intentional: source MD already passed this skill's own novelty + cross-model review. HTML render is a structural conversion, not a new claim-audit gate. Output lands at `idea-stage/IDEA_REPORT.html` with embedded source SHA256 + render timestamp.

**Non-blocking**: if `/render-html` fails (helper missing, Codex MCP unavailable, file write error), log the failure and continue — the HTML view is a convenience artif

…

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [wanshuiyin](https://github.com/wanshuiyin)
- **Source:** [wanshuiyin/Auto-claude-code-research-in-sleep](https://github.com/wanshuiyin/Auto-claude-code-research-in-sleep)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-wanshuiyin-auto-claude-code-research-in-sleep-idea-discovery
- Seller: https://agentstack.voostack.com/s/wanshuiyin
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.