idea-discovery

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.
• 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

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:

# 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"

# Otherwise (the common case), include gemini explicitly for broader discovery:
/research-lit "$ARGUMENTS" — sources: all, gemini

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"

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-5.4 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). 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-5.4 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]"

What this does:

• GPT-5.4 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-5.4 review (up to 5 rounds, until score ≥ 9)
• Generate a claim-driven experiment roadmap with ablations,

Content truncated.