$npx -y skills add PangenomeAI/academic-skills-food-nutrition --skill agri-pipelineMaster orchestrator for the whole agricultural research-to-publication workflow, run as a senior agricultural scientist. Coordinates the agri skills — journal selection, research (agri-research or agri-deep-research), writing and analysis (agri-paper), figures (food-figure), peer
| 1 | # Agri-Pipeline — Research-to-Publication for Agricultural Science |
| 2 | |
| 3 | **Run the `food-pipeline` skill exactly** — its subagents (`intake_router`, |
| 4 | `quality_gate`), its **stages 0–6**, its quality gates, its mode advisor, and its |
| 5 | review/revision authorization defaults — with the agriculture substitutions in |
| 6 | [`agri-research/references/agriculture-domain.md`](../agri-research/references/agriculture-domain.md). |
| 7 | Read that file first. No new machinery here. |
| 8 | |
| 9 | ## The substitutions |
| 10 | 1. **Persona** — a **senior agricultural scientist of the specific discipline**, |
| 11 | carried into every stage and every skill it dispatches (domain §2). |
| 12 | 2. **Route to the `agri-*` skills**, not the food ones: |
| 13 | |
| 14 | | Stage | Skill | |
| 15 | |---|---| |
| 16 | | 0 · ROUTE | `intake_router` + `journal-selector` (agriculture coverage map) | |
| 17 | | 1 · RESEARCH | **`agri-research`** (or **`agri-deep-research`**) | |
| 18 | | 2 · WRITE | **`agri-paper`** → figures via **`food-figure`** (domain-neutral) | |
| 19 | | 3 · REVIEW | **`agri-review`** | |
| 20 | | 4 · REVISE | **`agri-paper`** (revise) | |
| 21 | | 5 · RE-REVIEW | **`agri-review`** (re-review) | |
| 22 | | 6 · FINALIZE | **`agri-paper`** (format-convert) | |
| 23 | |
| 24 | 3. **Evidence base** — Tier 1 Q1/Q2 agriculture + Nature/Science/Cell/PNAS + Q1/Q2 |
| 25 | adjacent disciplines; Q3 for gaps; **Q4 avoided** (domain §3). |
| 26 | |
| 27 | `food-figure` is used directly and deliberately — it is domain-neutral and already |
| 28 | renders at the journal spec. |
| 29 | |
| 30 | ## Inherited unchanged (not optional) |
| 31 | - **Journal resolved once** at ROUTE via `journal-selector` and passed to every |
| 32 | downstream skill, so none re-asks. |
| 33 | - **Knowledge reuse:** when Stage 1 runs, its evidence base is carried into Stages |
| 34 | 3/5 so `agri-review` does **not** re-search the field — it reuses it and tops up |
| 35 | with `agri-research` **quick brief** key reviews read in full. |
| 36 | - **Review/revision defaults:** **one** review→revise round; a second round **and** |
| 37 | in-place Tracked Changes on the original Word file each require **explicit author |
| 38 | authorization**. |
| 39 | - **Deliverables — exactly two files, both `.docx`:** one **manuscript**, and **one |
| 40 | `Review_and_Response_Report_<slug>_<date>.docx`** carrying **both the reviewer |
| 41 | feedback and the editing response** — `agri-review` writes the feedback at Stage 3, |
| 42 | `agri-paper` fills each `Response (type)` into that **same file** at Stage 4, and an |
| 43 | authorized round 2 appends `R2-*` items to it. **No separate reviewer report, no |
| 44 | standalone response letter, never Markdown** |
| 45 | (`food-review/references/report-format.md`). |
| 46 | - Quality gates, anti-fabrication grounding, four-gate citations, privacy scan, |
| 47 | `human-writing.md`, and the **mandatory AI-use disclosure**. |
| 48 | |
| 49 | ## Agricultural gates (domain §5) |
| 50 | At the Stage-2 gate, do not pass a manuscript that omits **the experimental unit**, |
| 51 | field-trial reporting (site, season/years, soil, cultivar, design, replication), or |
| 52 | ethics/**ARRIVE** for animal work — these are the flaws that end agricultural papers |
| 53 | at review. |