grind-lab-core/night_owl_research_agent/paper-plan

Skill: paper-plan

You produce a concrete, actionable paper outline before any writing begins from: $ARGUMENTS

Constants

• REVIEWER_MODEL = `gpt-5.4` — Model used via Codex MCP for outline review. Must be an OpenAI model.
• TARGET_VENUE = `IJGIS` — Default venue. User can override (e.g., /paper-plan "topic" — venue: AAAG). Supported: TGIS, RSE, ISPRS_JPRS, AAG_ANNALS, IEEE_TGRS, ICML, ICLR, NeurIPS, CVPR, ACL, AAAI, ACM, IEEE_JOURNAL (IEEE Transactions / Letters), IEEE_CONF (IEEE conferences).
• MAX_PAGES = 40 — Adjust based on paper template selected. ML conferences typically 8–10 pages (excluding refs/appendix). IEEE venues include references in page count.
• MAX_PRIMARY_CLAIMS = 2 — One dominant contribution + one supporting. Prevents scope creep.
• MAX_FIGURES = 8 — Soft cap; hero figure + up to 7 supporting figures/tables.
• OUTPUT_PATH = `output/PAPER_PLAN.md`

Orchestra-Guided Writing Overlay

Keep the existing workflow and outputs, but use the shared references below to improve the quality of the story and outline.

• Read skills/knowledge/academic-writing.md when framing the one-sentence contribution, Abstract, Introduction, Related Work, or hero figure.
• Read skills/knowledge/geoai-domain.md for GIScience/GeoAI framing conventions.
• Read skills/knowledge/spatial-methods.md when planning methodology sections for spatial analysis papers.
• Only load these references when needed; do not paste their full contents into the working draft.

Phase 0: Load Checkpoint (Resume Support)

Check for existing state:

1. Read output/PAPER_PLAN.md — if it exists and was generated in the current session, ask the user whether to rebuild or refine it.
2. Read handoff.json — if pipeline.stage indicates paper-plan is in progress, resume from the last recorded phase.

If no checkpoint exists, proceed to Phase 1.

Phase 1: Gather Context

Read all available upstream artifacts. Each file is optional — missing files are soft failures. Record which files were found and which were missing; this information goes into the plan's "Input Files Used" and "Missing Inputs" sections.

Primary Inputs (read in this order)

Context Synthesis Rules

1. NARRATIVE_REPORT.md is king — if it exists, it is the primary planning source. It was specifically designed to consolidate all upstream artifacts for the paper-writing pipeline. Use other files only to fill gaps or verify claims.
2. RESEARCH_PLAN.md is the research intent — it defines what the researcher wants to achieve. FINAL_PROPOSAL.md is the refined version. EXPERIMENT_RESULT.md is what actually happened. The outline must reconcile all three.
3. Never fabricate from gaps — if a file is missing, the outline must explicitly mark the affected sections as [NEEDS: <missing-file>] rather than inventing content.
4. Numbers are sacred — copy metrics verbatim from EXPERIMENT_RESULT.md and APPROVED_CLAIMS.md. Never round, paraphrase, or extrapolate.

Phase 2: Determine Venue and Paper Type

Step 2.1: Resolve TARGET_VENUE

Check in this order (first match wins):

1. Explicit user argument (e.g., — venue: ISPRS_JPRS)
2. RESEARCH_PLAN.md target venue field
3. NARRATIVE_REPORT.md venue target section
4. output/PAPER_PLAN.md existing venue (if refining)
5. Default: IJGIS

Step 2.2: Load Venue Template

Based on TARGET_VENUE, choose the appropriate template from templates/:

If the template does not exist locally, use WebSearch to retrieve the venue's author guidelines and page limits.

Step 2.3: Determine Paper Type

Infer from the contributions and method:

Step 2.4: Set Page Budget

Derive from venue:

Phase 3: Build the Outline

Write output/PAPER_PLAN.md following the structure in templates/PAPER_PLAN_TEMPLATE.md (Sections §0–§26). The template is the target schema; fill every section with content derived from the upstream artifacts.

Section-by-Section Generation Rules

§0 Document Status — Fill version, date, venue, manuscript type, readiness level, list of input files consumed and missing.

§1 One-Paragraph Summary — Synthesize from NARRATIVE_REPORT.md §1 or FINAL_PROPOSAL.md contributions. The one-sentence claim must be specific, defensible, and evidence-based. Draft a 150–250 word abstract-style summary.

§2 Target Journal Strategy — Use venue template to fill journal fit, audience, expectations. Read skills/knowledge/academic-writing.md for framing advice.

§3 Research Context and Motivation — Draw from LIT_REVIEW_REPORT.md synthesis + RESEARCH_PLAN.md problem statement. Quantify the problem scale.

§4 Research Gap — Extract directly from LIT_REVIEW_REPORT.md Gap Analysis section. List specific gaps with boundary citations. Explain why existing work is insufficient.

§5 Novelty and Contributions — Numbered contributions from FINAL_PROPOSAL.md or NARRATIVE_REPORT.md. Each contribution tied to a specific experiment/claim. Include "What This Paper Is Not Claiming" to set expectations.

§6 Research Questions and Hypotheses — From RESEARCH_PLAN.md or FINAL_PROPOSAL.md. Align RQs with experiment design from EXPERIMENT_PLAN.md.

§7 Study Scope and Boundaries — Spatial, temporal, data, and method scope. Explicit limitations of scope.

§8 Data and Materials — From DATA_MANIFEST.md and EXPERIMENT_PLAN.md. For each dataset: source, resolution, temporal coverage, preprocessing, license.

§9 Methodological Plan — From FINAL_PROPOSAL.md method description and EXPERIMENT_PLAN.md design. Include baselines, evaluation protocol, spatial analysis components. Read skills/knowledge/spatial-methods.md for GIScience method framing.

§10 Experiments — From EXPERIMENT_PLAN.md (design) and EXPERIMENT_RESULT.md (outcomes). Mark incomplete experiments as [PENDING].

§11 Results Summary — From EXPERIMENT_RESULT.md and APPROVED_CLAIMS.md. Headline findings in priority order. Quantitative results with exact values. Flag missing results as [NEEDS: experiment completion].

§12 Claim-to-Evidence Map — The backbone of the plan. Every major claim maps to: evidence source, quantitative support, figure/table ID, experiment ID, confidence level. Unsupported claims go into "Unsupported or Weak Claims" subsection.

§13 Figures Plan — From NARRATIVE_REPORT.md figure plan or EXPERIMENT_RESULT.md outputs. For each figure: ID, type, description, data source, status, priority.

CRITICAL: Hero Figure (Fig. 1) — Describe in detail:

- What methods/concepts are being compared

- What the visual difference should demonstrate

- Caption draft that clearly states the comparison

- Why a skim reader understands the paper from this figure alone

§14 Tables Plan — Required tables: dataset summary, baseline comparison, ablation (if applicable), hyperparameters, error analysis.

§15 Related Work Synthesis — From LIT_REVIEW_REPORT.md thematic synthesis. Group into 3–4 clusters that map to the paper's related work section. For each cluster: summary, representative studies, how our work relates and differs.

§16 Discussion Plan — Interpretation themes, implications (GIScience/GeoAI + practical), responsible research considerations, generalizability.

§17 Limitations and Future Work — Concrete limitations from EXPERIMENT_RESULT.md and AUTO_REVIEW_REPORT.md. Severity assessment. 3–5 specific future work directions.

§18 Reproducibility and Open Science Plan — Code availability, data availability, reproducibility assets checklist.

§19 Manuscript Structure Plan — Section outline (5–8 sections, flexible) with word budgets. Per-section: goal, key points, gap statement, contributions paragraph.

§20 Abstract Blueprint — Sentence-by-sentence abstract structure: background → gap → method → data → main results → significance.

§21 Title and Framing Options — 3 candidate titles. Dominant framing (methodological / applied / benchmark / conceptual).

§22 Citation and Evidence Bank — Per-section citation plan from LIT_REVIEW_REPORT.md verified citations. Flag unverified citations with [VERIFY].

§23 Writing Instructions for Downstream Agent — Non-negotiable writing goals, style instructions, section priorities, writing risks to avoid. Venue-specific tone guidance.

§24 Open Issues Before Drafting — Critical gaps, nice-to-have improvements, required follow-up actions with owners and priorities.

§25 Final Readiness Assessment — Ready for full/partial/skeleton draft? Minimum conditions for drafting. Recommended drafting strategy.

§26 Executive Summary for Manuscript Writer — Concise: what the paper is about, why publishable, strongest/weakest evidence, what to emphasize, what to be careful about.

Section Count and Word Budget

IMPORTANT: The section count is FLEXIBLE (5–8 sections). Choose what fits the content and paper type best. The template sections above are the planning schema — the actual manuscript sections in §19 are determined by venue and paper type.

Example word budgets for a 25-page IJGIS paper (~8000 words):