idea-creator

Research Idea Creator

Generate publishable research ideas for: $ARGUMENTS

Overview

Given a broad research direction from the user, systematically generate, validate, and rank concrete research ideas. This skill composes with /research-lit, /novelty-check, and /research-review to form a complete idea discovery pipeline.

Constants

• PILOT_MAX_HOURS = 2 — Skip any pilot estimated to take > 2 hours per GPU. Flag as "needs manual pilot".
• PILOT_TIMEOUT_HOURS = 3 — Hard timeout: kill pilots exceeding 3 hours. Collect partial results if available.
• MAX_PILOT_IDEAS = 3 — Pilot at most 3 ideas in parallel. Additional ideas are validated on paper only.
• MAX_TOTAL_GPU_HOURS = 8 — Total GPU budget for all pilots combined.
• REVIEWER_MODEL = gpt-5.5 — Default model for the Codex backend. Must be an OpenAI model (e.g., gpt-5.5, o3, gpt-4o). Manual backend uses whatever model the user chooses, but it must be a non-Claude model — the executor is Claude, so pasting into any Claude product makes Claude judge Claude and voids the cross-model invariant (see shared-references/reviewer-routing.md).
• REVIEWER_BACKEND = codex — Default: Codex MCP (xhigh). Override with — reviewer: oracle-pro for Oracle MCP, or — reviewer: manual for Manual Review MCP. If manual-review MCP is unavailable, stop and print the install command; do not fall back to Codex. See shared-references/reviewer-routing.md.
• OUTPUT_DIR = idea-stage/ — All idea-stage outputs go here. Create the directory if it doesn't exist.

💡 Override via argument, e.g., /idea-creator "topic" — pilot budget: 4h per idea, 20h total.

Reviewer Calling Convention

When calling the reviewer for idea evaluation, branch on REVIEWER_BACKEND:

If REVIEWER_BACKEND = codex: Use mcp__codex__codex for new review threads. Use mcp__codex__codex-reply for follow-up rounds (reuse threadId).

If REVIEWER_BACKEND = manual: Use mcp__manual_review__review for new review threads with: prompt: [exact same prompt that would go to Codex] config: {"model_reasoning_effort": "xhigh"} Save the returned threadId. Use mcp__manual_review__review_reply for follow-up rounds with: threadId: [saved manual-review threadId] prompt: [follow-up prompt] config: {"model_reasoning_effort": "xhigh"}

Prompt fidelity: the manual prompt must be exactly the same text that Codex would receive. Review tracing applies equally to both backends.

Workflow

Phase 0: Load Research Wiki (if active)

Skip this phase entirely if research-wiki/ does not exist.

If research-wiki/ exists, resolve the canonical helper using the shared resolution chain (see ../research-wiki/SKILL.md for the contract):

cd "$(git rev-parse --show-toplevel 2>/dev/null || pwd)" || exit 1
ARIS_REPO="${ARIS_REPO:-$(awk -F'\t' '$1=="repo_root"{print $2; exit}' .aris/installed-skills.txt 2>/dev/null)}"
WIKI_SCRIPT=".aris/tools/research_wiki.py"
[ -f "$WIKI_SCRIPT" ] || WIKI_SCRIPT="tools/research_wiki.py"
[ -f "$WIKI_SCRIPT" ] || { [ -n "${ARIS_REPO:-}" ] && WIKI_SCRIPT="$ARIS_REPO/tools/research_wiki.py"; }
[ -f "$WIKI_SCRIPT" ] || {
  echo "WARN: research_wiki.py not found at .aris/tools/, tools/, or \$ARIS_REPO/tools/." >&2
  echo "      The idea-creation primary output (idea ranking) will still be produced." >&2
  echo "      Wiki integration (load query_pack, write idea pages, add edges, rebuild query_pack) will be skipped." >&2
  echo "      Fix: rerun 'bash tools/install_aris.sh', export ARIS_REPO, or 'cp <ARIS-repo>/tools/research_wiki.py tools/'." >&2
  WIKI_SCRIPT=""
}

if research-wiki/query_pack.md exists AND is less than 7 days old:
    Read query_pack.md and use it as initial landscape context:
    - Treat listed gaps as priority search seeds
    - Treat failed ideas as a banlist (do NOT regenerate similar ideas)
    - Treat top papers as known prior work (do not re-search them)
    Still run Phase 1 below for papers from the last 3-6 months (wiki may be stale)
else if research-wiki/ exists but query_pack.md is stale or missing:
    if [ -n "$WIKI_SCRIPT" ]: python3 "$WIKI_SCRIPT" rebuild_query_pack research-wiki/
    Then read query_pack.md as above

Phase 1: Landscape Survey (5-10 min)

Map the research area to understand what exists and where the gaps are.

1. Scan local paper library first: Check papers/ and literature/ in the project directory for existing PDFs. Read first 3 pages of relevant papers to build a baseline understanding before searching online. This avoids re-discovering what the user already knows.

1. Search recent literature using WebSearch:
2. Top venues in the last 2 years (NeurIPS, ICML, ICLR, ACL, EMNLP, etc.)
3. Recent arXiv preprints (last 6 months)
4. Use 5+ different query formulations
5. Read abstracts and introductions of the top 10-15 papers

1. Build a landscape map:
2. Group papers by sub-direction / approach
3. Identify what has been tried and what hasn't
4. Note recurring limitations mentioned in "Future Work" sections
5. Flag any open problems explicitly stated by multiple papers

1. Identify structural gaps:
2. Methods that work in domain A but haven't been tried in domain B
3. Contradictory findings between papers (opportunity for resolution)
4. Assumptions that everyone makes but nobody has tested
5. Scaling regimes that haven't been explored
6. Diagnostic questions that nobody has asked

Phase 1.5: Parallel lens fan-out (Tier-aware) — breadth, not verdict

Idea generation benefits from breadth: more independent analytic angles surface more candidate ideas. This skill fans out candidate generation across analytic lenses, then funnels every candidate through the single Phase-4 cross-model jury. Fan-out widens the jury's input; it never makes the accept/reject decision. This follows [shared-references/fan-out-pattern.md](../shared-references/fan-out-pattern.md); the verdict stays cross-model per [shared-references/acceptance-gate.md](../shared-references/acceptance-gate.md) (idea novelty/quality is a Type-B verdict — same-family generation is fine, same-family acquittal is not).

Lenses (the structural-gap angles from Phase 1, step 3): method-transfer (works in domain A, untried in B) · contradiction (conflicting findings to resolve) · untested-assumption (everyone assumes, nobody tested) · scaling-regime (unexplored regime) · diagnostic (question nobody asked). This set is a floor, not a ceiling — add a domain-specific lens when the direction warrants.

Tier-portable dispatch (the Phase-4 jury downstream is identical on every tier):

• Tier 1 (Workflow available): spawn one Claude subagent per lens;

each runs the Phase-1 survey through its lens and the Phase-2 generation prompt restricted to that lens, returning candidates as structured output.

• Tier 2 (Agent tool, no Workflow): spawn the same per-lens subagents via

the Agent tool.

• Tier 3 (no spawning): enumerate the lenses sequentially in one pass —

the original single-thread behavior, made explicit. No capability assumed.

Why the lens shards are Claude, not Codex. Generation is candidate

production, not a verdict, so same-family is safe — and Codex MCP is

serial (concurrent codex calls hang), so spending its scarce capacity

on parallel generation is both unsafe-to-parallelize and wasteful. Reserve

Codex for the one Phase-4 jury call. On Tier 1/2 the lens subagents are the

generators; the single Phase-2 codex brainstorm below still runs once as an

optional cross-model seed (a generator, not a judge), and its ideas join

the merged pool.

Per-shard output (the generation-fan-out schema from [fan-out-pattern.md](../shared-references/fan-out-pattern.md) — shard_id + candidates[] + per-item dedup_key):

{"shard_id": "<lens id>", "candidates": [{"summary": "...", "hypothesis"