More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Modelling Paper: From Model Results to a Full Draft

    1

    This skill makes your agent a co-author in two roles.

    $12

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    Modelling Paper: From Model Results to a Full Draft

    Modelling Paper: From Model Results to a Full Draft

    Example session with this skill installed

    Help me turn these model results into a paper. My model code is model.py, the results are in research/results/ and my rough outline is research/outline.md. Start with the Methods: the full model, the notation table and the stability analysis, and check the maths. Then tell me what reviewers will attack. Go ahead without waiting for my confirmation.

    • Read your context and instructions
    • Compiled the modelling paper

    I've written the Methods and checked the maths numerically. Two things in your outline don't survive the check.

    What I wrote: research/paper/methods.md (the full model, notation table with sources, stability analysis, numerical settings), research/paper-notes.md (checks, open questions, reviewer objections R1-R11), research/claims.md and a Figure 1 plan. I added a proposed one-sentence claim to the top of your outline and left your notes untouched.

    The maths, checked:

    The threshold has a closed form. Runaway starts at a saddle-node (fold), where the slope of self-heating equals 1/R_tot: T_a,crit = T* - R_tot P_J - R T*^2/E_a (Eq. 9). It matches your bisection to within 3e-6 K for every R_out.

    All four result files reproduce from model.py. The analytic Jacobian matches a finite-difference one to 4e-10, and the time to runaway agrees across three ODE solvers.

    • Your dashed curve is a saddle: the runaway threshold for perturbations.

    Fix before anything else:
    1.

    No Hopf bifurcation is possible (outline result 3). Both off-diagonal Jacobian entries are positive everywhere, so the eigenvalues are always real. The testable early warning is critical slowing down: the slow eigenvalue falls from -3.1e-3 to -4e-4 per second as the threshold approaches.
    2. "Safe below T_crit" is local. 1 K below T_crit, a core that starts 10 K hot still runs away (8 K does not).
    3.

    The current doesn't match. Your outline says 2.5 A; the code and every result use 3.0 A. With 2.5 A the threshold is 57.49 C instead of 56.87 C. Which is right?
    4.

    make_inputs.py is missing. My re-implementation of the Monte Carlo is consistent with your numbers but not an exact reproduction.

    What reviewers will attack: the unsourced self-heating rate, which drives most of the uncertainty (rank correlation -0.92) and is sampled independently of the activation energy; constant current for hours (a 2.5 Ah cell at 3 A is empty in 50 minutes); the unjustified two-node lumping; natural convection as a guess; no validation; novelty against classical Semenov theory.

    Still open: 13 [CITE] and [CHECK] placeholders, including your two "need to find the ref" sources. Should I run a literature search, or do you have the papers?

    Connects securely to your tools. The creator never sees your data.

    About this skill

    What it does

    A modelling paper is judged on its mathematics: is the model written out in full, is every parameter sourced, do the derived results hold, and does the text claim more than the model shows? This skill makes your agent a co-author in two roles. As PI it owns the story, pins the one-sentence claim before anything is drafted, and asks "what will Reviewer 2 say?" before anyone else does. As postdoc it writes the full prose and the complete Methods mathematics, then checks it.

    • Starts from what you have. It reads your existing files first: model code, results, outline, partial draft. The Methods describe what your code actually solves, and any difference between the code and your notes is reported.
    • The full Methods mathematics. The complete model, every symbol defined, initial and boundary conditions, scaling, derived results (steady states, stability conditions, thresholds, distributions) with enough derivation to check, and the numerical settings. A notation table gives each symbol's meaning, units, value or range, and source.
    • Checked, not asserted. Dimensions, limiting cases, the code against the equations, and a numerical check of every analytic result when Python is available. Every number in the prose comes from a computation. Claims in your notes that the model cannot support are flagged before they are written up.
    • Any model class, any field. A per-class checklist covers deterministic ODEs, DDEs and PDEs, master equations and SDEs, agent-based models, statistical and inference models, networks, optimisation and control, numerical-methods papers and surrogates. Domain packs add the vocabulary, conventions and field-specific objections for biology, epidemiology, ecology, and physics and engineering.
    • Reviewer objections, rated and planned. A running list in your notes: each objection specific to your paper, rated fatal, major or minor, with the planned answer, drawn from a general catalogue (model construction, parameters, identifiability, fitting, numerics, robustness, scope of claims, novelty) plus the domain pack.
    • No invented references. A citation is either verified (seen in a search result, supplied by you, or read) or marked. A parameter with no known source gets a [CITE] marker, never a plausible-looking citation.

    In the demo

    Given a synthetic two-node thermal model of a battery cell, its code, four result files and a rough outline, the agent wrote the Methods with a closed-form runaway threshold that matched the user's bisection to 3e-6 K. It proved that the Hopf bifurcation the outline hoped for cannot occur (the system is cooperative, so its eigenvalues are always real). It narrowed "safe below the threshold" to a local result and caught that the outline said 2.5 A while the code used 3.0 A. It left the bibliography empty because no sources had been given, and listed eleven reviewer objections.

    What's included

    • SKILL.md: the two roles, session start, project layout, Methods rules, figure planning, citations, claims, reviewer objections, strategic questions, export
    • references/methods-maths.md: what every model needs, what each model class adds, the verification pass, a Methods skeleton
    • references/reviewer-objections.md: the general objection catalogue
    • references/literature-and-citations.md: where to search, when a reference counts as verified, how to mark the rest
    • references/domains/: biology, epidemiology, ecology, physics and engineering
    • scripts/scaffold_paper.py (creates the paper skeleton, never overwrites) and scripts/check_refs.py (citation audit)

    Part of the Aurora Lab research pipeline

    Works entirely on its own. With a research/ folder it keeps the manuscript in research/paper/ and its working notes beside it. Pairs well with Aurora Lab's Claim Check, Scientific Voice and Journal House Style, but needs none of them.

    It writes and checks the paper; it does not run your study. Numerical checks need a working environment for your model (for example Python with numpy and scipy); without one, the agent checks by hand and lists what it could not verify. The checks are a careful review, not a formal proof. It cites only references it has verified, so without a literature-search tool or your own PDFs and BibTeX, references stay as [CITE] placeholders for you to fill. Domain packs cover biology, epidemiology, ecology, and physics and engineering; in other fields it works from the general core and asks you about your community's conventions. It does not produce figures (it plans them and writes captions), format for a specific journal, or edit sentence-level style. Markdown is the working format; LaTeX export is a separate step on request.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 28 days ago

    Requires

    Python

    Permissions

    Terminal / Shell
    Read Files
    Write Files

    File Scopes

    research/outline.md
    research/paper-notes.md
    research/paper/**
    research/figures/*.md
    research/claims.md
    research/refs.bib
    research/results/*.json
    **/*.py

    Terminal: runs the two bundled scripts (paper skeleton, citation audit) and short throwaway scripts that check the maths numerically, kept in the system temp directory unless you ask to keep them. Reads your model code (any file or folder you point it to; Python shown as **/*.py), research/results/*.json, and the existing paper files. Writes the manuscript sections in research/paper/, the working files research/outline.md and research/paper-notes.md, new figure plans in research/figures/, its lines in research/claims.md and entries in research/refs.bib. It never edits your code, data or result files. The scripts use no network. A literature search happens only through a search tool your agent already has, and only when you agree to it.

    Listed28 days ago

    What's inside

    Frequently Asked Questions

    Any paper whose contribution rests on a mathematical or computational model: deterministic ODEs, DDEs and PDEs, master equations and SDEs, agent-based models, statistical and inference models, network models, optimisation and control, numerical-methods papers and surrogate models. Each class has a checklist of what its Methods must contain.

    Yes. The core is field-neutral, and the demo is an engineering model: a thermal model of a battery cell. Domain packs add vocabulary and field-specific reviewer objections for biology, epidemiology, ecology, and physics and engineering. For other fields, such as economics, climate or neuroscience, it works from the core and asks you which conventions and reporting standards your community expects.

    It runs the checks as part of the workflow: units, limiting cases, your code against the written equations, and a numerical check of every analytic steady state, threshold or distribution when a Python environment is available. It records what it checked, how, and what it could not check in the notes. In the demo it verified a closed-form threshold against the user's code to 3e-6 K and showed that a Hopf bifurcation claimed in the outline is impossible.

    No. A reference is cited only if the agent saw it in a search result with a DOI or stable identifier, you supplied it, or it read the paper. Anything else is marked UNVERIFIED or left as a [CITE: what is needed] placeholder. The bundled check_refs.py lists every placeholder and unverified key before export. In the demo the bibliography stayed empty because the user's notes named no sources.

    Both. It drafts every section (Introduction, Results, Discussion, Methods, and the abstract last) and starts wherever you ask. The demo asked for the Methods and the reviewer objections first, so that is what it produced, plus a Figure 1 plan and a proposed one-sentence claim.

    Whatever you have: model code, result files, an outline, a partial draft or just the idea. It reads existing files before asking anything. For citations, a literature-search tool your agent already has, or your own PDFs, BibTeX export or DOIs.

    This skill co-writes the paper and its mathematics. Claim Check audits finished prose sentence by sentence against the evidence. They pair well, and neither needs the other.

    New in Writing & Content