neuroflow

Infinite improvement loop for any research artifact — point it at any file(s), and it runs a worker-evaluator loop indefinitely: one focused change per iteration, keep or revert based on relative comparison, dashboard at localhost:8765. Never stops until interrupted.

Assemble a computational brain model or continue working on one — neuron models, network topology, connectivity, and simulation framework setup.

Run a parameter search or fit a computational brain model to experimental data.

Run a computational brain model simulation — configure run parameters, launch the simulation, and collect outputs.

Run an analysis pipeline on preprocessed data — ERPs, time-frequency, connectivity, decoding, GLM, or other modality-appropriate analyses.

Run a preprocessing pipeline on ingested data — filtering, ICA, epoch creation, artifact rejection, and QC.

Data intake — locate data, validate BIDS structure, and run conversion scripts to get raw data ready for preprocessing.

Full experiment preparation — paradigm design (PsychoPy), recording setup, and instrument configuration.

Log user dissatisfaction — record what went wrong in core behavior, science quality, or interaction experience. Optionally prepares a GitHub issue report for the neuroflow team.

Manage grant documents and track project expenses. Covers budget planning, expense logging, financial reporting to funders, and grant compliance.

Personal research OS — link a private GitHub repository to store your research profile, cross-project Kanban task board, and project registry with phase tracking. Supports profile creation, task management, project registry, GitHub sync, phase auto-tracking, and credential export.

Context-aware git utility — reads the current repo state and suggests or executes the most appropriate git action. Supports shorthand aliases (p, pl, ps, a, c, ac, acp, b, pr) so you can commit, push, pull, branch, or open a PR with minimal typing.

Write a full grant application. Starts with an interactive interview, discovers ideation outputs, accepts funding call documents or URLs, adapts to any funder (NIH, ERC, Wellcome, GAČR, etc.), and drafts section by section with word-count tracking.

Connect neuroflow to a shared team GitHub org repo for research direction coordination, explicit knowledge sharing, and team-aware recommendations. Use --init to connect, --sync to pull updates, --view to see team state, --share to explicitly push a finding, --recommend to get team-aware suggestions for the current phase. Never auto-shares personal project data.

The very beginning of a research project — brainstorm a research question, explore literature inline (no sub-agents), formalize an existing idea into a project definition, or produce a project proposal document.

Personal support companion — talk about how you feel, share what's on your mind, or just decompress. A quiet moment in the research workflow for when you're overwhelmed, burned out, or just need to think out loud.

Interview preparation from either side of the table — determine whether you are the interviewer or interviewee, generate tailored questions grounded in your research context, run a practice Q&A, and optionally evaluate answers.

Main entry point for a neuroflow project. If .neuroflow/ exists, shows current phase and status. If not, interviews the user and creates the .neuroflow/ folder structure.

Lightweight live note-taking — capture notes during a meeting, talk, or session, then reformat them into a clean structured document.

Output project memory or the whole project — pack it as a zip archive or copy it to a target location for sharing, archiving, or handoff.

Unified manuscript writing and review command — drafts a neuroscience paper section by section, then runs every draft through a brutal paper-writer → paper-critic loop before saving anything.

Show the current project phase and all phases worked on so far. Optionally switch to a different phase.

Define and run a multi-step research pipeline across any sequence of neuroflow phases. Interactive by default — pauses between steps for approval. Pass --executor to run in brutal mode with no stops.

Generate a LaTeX academic conference poster from project memory. Selects a template (A0/A1/A2, portrait/landscape, custom sizes), injects project content, adds a QR code, and runs the result through the poster-critic agent in an iterative review loop.

Create and manage pre-registration documents for OSF or AsPredicted. Covers study design, hypotheses, analysis plan, and linking registered reports.

Neuroscience quiz — test your knowledge with flashcards, a pub quiz, or rapid-fire questions. Covers general neuroscience or a chosen subfield.

Peer review a colleague's neuroscience paper — structured referee report with major/minor issues, calibrated to the target journal.

Lightweight search across the project or neuroflow memory. Use a tag to scope the search — `memory` searches inside .neuroflow/ only, `project` searches the codebase outside .neuroflow/. Returns a fast summary of what was found.

Audit of .neuroflow/ — checks flow.md completeness, timestamps, broken references, preregistration drift, session consistency, and personal sensitive information (emails, passwords, private keys, names, institutions). Scoped to .neuroflow/ by default; asks before scanning the full workspace. Writes a report to .neuroflow/sentinel.md.

Interactive credential wizard for neuroflow MCP integrations. Checks Miro, Google Workspace CLI credentials, and custom LLM provider settings, prompts for missing values, and saves them to .neuroflow/integrations.json (per-project) or ~/.neuroflow/integrations.json (global, device-wide).

Build a presentation from selected areas of the project — pick phases, figures, and key findings, then get a structured slide deck ready to export.

Build a lab tool or software pipeline — real-time systems, data acquisition, LSL integrations, custom analysis pipelines, or other technical infrastructure.

Create a comprehensive testing pipeline to verify that a tool or paradigm works correctly — timing, markers, data output, edge cases.

Generate a structured report from .neuroflow/ contents — for a specific phase, multiple phases, or the whole project.

Autoresearch loop agent — runs the infinite worker-evaluator improvement loop for any phase. Makes one focused change per iteration via a worker, compares to previous best via an evaluator, keeps or reverts. Never stops until the human interrupts. Inspired by Andrej Karpathy's autoresearch (MIT).

Personal identity agent — reads the user's flowie profile and generates deeply personalized assistance based on their research stances, writing style, methodological preferences, and intellectual fingerprint. Invoked when Claude needs to tailor its output to who the user actually is.

Literature review specialist. Runs 12 sequential analytical lenses on a set of downloaded papers — from landscape mapping to future research agenda — using the worker-critic loop to ensure rigour. Scoped to the ideation phase.

Hyper-critical manuscript reviewer for the unified paper phase. Applies the full six-area neuroflow:review-neuro methodology to every section draft — as if reviewing for Nature Neuroscience or Neuron. Returns [STATUS: APPROVED] or [STATUS: REJECTED] with specific, actionable feedback. Never produces content.

Specialist manuscript writer for the unified paper phase. Drafts neuroscience manuscript sections from upstream project memory, journal guidelines, and analysis results — operates inside a brutal write→critique loop with the paper-critic agent and revises until approved or the loop is exhausted.

Hyper-critical academic conference poster reviewer. Evaluates LaTeX poster `.tex` source against design, content, and communication standards. Returns [STATUS: APPROVED] or [STATUS: REJECTED] with specific, actionable feedback. Used by /poster in the iterative worker-critic loop (max 3 cycles). Never produces content — only audits.

Standalone academic paper research specialist. Searches both PubMed and bioRxiv for a given topic, returns a clean structured list of results, and supports follow-up actions (download full text, save as markdown, deeper synthesis). NOTE: The /ideation command and ideation agent perform searches inline — they do NOT spawn this agent as a sub-agent. Use this agent only for ad-hoc literature searches outside the ideation workflow.

Plugin development coherence guard. Monitors consistency of the neuroflow plugin itself — folder names vs frontmatter, version sync (plugin.json, README, mkdocs.yml), README tables, docs website navigation, dead references inside SKILL.md files, naming overlaps between skills and agents, and personal sensitive information (emails, passwords, private keys, names, institutions).

Project coherence guard. Audits .neuroflow/ for internal consistency — checks flow.md completeness, timestamps, broken references, preregistration drift, session consistency, and personal sensitive information (emails, passwords, private keys, names, institutions). Scoped to .neuroflow/ by default; full workspace scan is opt-in. Called by the /sentinel command.

Infinite improvement loop for any research artifact in any phase — worker makes one focused change per iteration, evaluator compares to previous best, keeps or reverts. Loop never stops until the human interrupts it. Inspired by Andrej Karpathy's autoresearch (MIT).

Remove AI writing signatures from prose. Makes text sound genuinely human-authored — varied rhythm, natural register, no AI tells. Use when drafting, editing, or reviewing any text that needs to read as if a person wrote it.

Core rules and lifecycle for all neuroflow commands and agents. Read this whenever running any neuroflow command to understand the .neuroflow/ folder structure and the required behaviour at the start and end of every command.

Guide for developing and maintaining the neuroflow plugin. Use when adding skills, commands, agents, or hooks to neuroflow, bumping the version, or publishing updates to GitHub.

Complete API for Google NotebookLM — full programmatic access including features not in the web UI. Create notebooks, add sources (URLs, YouTube, PDFs, audio, video, images), generate all artifact types (podcast, video, slide deck, infographic, report, quiz, flashcards, mind map), and download results in multiple formats. Activates on explicit /notebooklm or intent like "create a podcast about X", "turn this into an audio overview", "generate a quiz from my research", "make an infographic", or "create a slide deck".

Phase guidance for the neuroflow /brain-build command. Loaded automatically when /brain-build is invoked to orient agent behavior, relevant skills, and workflow hints for assembling computational brain models.

Phase guidance for the neuroflow /brain-optimize command. Loaded automatically when /brain-optimize is invoked to orient agent behavior, relevant skills, and workflow hints for parameter search and model fitting.

Phase guidance for the neuroflow /brain-run command. Loaded automatically when /brain-run is invoked to orient agent behavior, relevant skills, and workflow hints for running brain model simulations.

Phase guidance for the neuroflow /data-analyze command. Loaded automatically when /data-analyze is invoked to orient agent behavior, relevant skills, and workflow hints for the data-analyze phase.

Phase guidance for the neuroflow /data-preprocess command. Loaded automatically when /data-preprocess is invoked to orient agent behavior, relevant skills, and workflow hints for the data-preprocess phase.

Phase guidance for the neuroflow /data command. Loaded automatically when /data is invoked to orient agent behavior, relevant skills, and workflow hints for the data intake phase.

Phase guidance for the neuroflow /experiment command. Loaded automatically when /experiment is invoked to orient agent behavior, relevant skills, and workflow hints for the experiment phase.

Phase guidance for the neuroflow /fails command. Orients agent approach for logging user dissatisfaction, categorising complaints accurately, and preparing GitHub issue reports.

Phase guidance for the neuroflow /finance command. Loaded automatically when /finance is invoked to orient agent behavior, relevant skills, and workflow hints for grant document management and expense tracking.

Phase guidance for the neuroflow /flowie command. Covers how to read and use the flowie profile for personalization, write rules for .neuroflow/flowie/, GitHub sync protocol, and profile-aware assistance across all phases.

Phase guidance for /git — context-aware git shorthand rules, smart push/pull logic, commit message generation, branch management, and PR creation workflow.

Phase guidance for /grant-proposal with deep funder knowledge, neuroscience-specific grant tactics, review criteria alignment, and workflow orchestration.

Team-level knowledge layer for neuroflow. Connects a researcher's personal neuroflow project to a shared GitHub organisation repo where team research directions, cross-project findings, and recommended methods are coordinated. Use when the user wants to sync their work with a team, share findings explicitly, view team directions, or get team-aware recommendations. Never automatically copies personal project data to Hive — all sharing is explicit.

Phase guidance for the neuroflow /ideation command. Loaded automatically when /ideation is invoked to orient agent behavior, relevant skills, and workflow hints for the ideation phase.

Phase guidance for the neuroflow /notes command. Loaded automatically when /notes is invoked to orient agent behavior, relevant skills, and workflow hints for the notes phase.

Phase guidance for the neuroflow /output command. Orients agent approach for outputting project memory or the whole project safely and with correct exclusions.

Phase guidance for the neuroflow /paper command. Loaded automatically when /paper is invoked to orient agent behavior, relevant skills, and workflow for the unified paper phase — covering manuscript drafting and rigorous internal peer review in a single write→critique loop.

Phase guidance for the neuroflow /pipeline command. Loaded automatically when /pipeline is invoked to orient agent behavior for planning and executing multi-step research pipelines.

Phase guidance for the neuroflow /poster command. LaTeX academic conference poster generation — template selection (A0/A1/A2, portrait/landscape, custom conference sizes), QR code integration, content extraction from project memory, and iterative poster-critic review loop.

Phase guidance for the neuroflow /preregistration command. Loaded automatically when /preregistration is invoked to orient agent behavior, relevant skills, and workflow hints for creating and managing pre-registration documents.

Phase guidance for the neuroflow /quiz command. Loaded automatically when /quiz is invoked to orient behaviour, question quality, and mode-specific workflow for neuroscience quiz sessions.

Phase guidance for the neuroflow /review command. Loaded automatically when /review is invoked. Orients Claude for acting as a formal peer reviewer of a colleague's neuroscience paper.

Lightweight search skill for the /search command. Instructs the agent to use flow.md files as a fast index, scope the search by tag (project or memory), and return a concise summary to the main agent.

Phase guidance for the neuroflow /slideshow command. Loaded automatically when /slideshow is invoked to orient agent behavior, slide structure decisions, format selection, and audience calibration for presentation generation.

Phase guidance for the neuroflow /tool-build command. Loaded automatically when /tool-build is invoked to orient agent behavior, relevant skills, and workflow hints for the tool-build phase.

Phase guidance for the neuroflow /tool-validate command. Loaded automatically when /tool-validate is invoked to orient agent behavior, relevant skills, and workflow hints for the tool-validate phase.

Phase guidance for the neuroflow /write-report command. Loaded automatically when /write-report is invoked to orient agent behavior, relevant skills, and workflow hints for the write-report phase.

Connect to Pupil Labs Neon eye-tracking glasses and collect real-time data streams (video, gaze, IMU, events). Use when working with Neon hardware via the Real-time API for live data visualization, data collection, or eye-tracking applications requiring device discovery, multi-threaded streaming, or hardware troubleshooting.

Run a rigorous pre-submission peer review of a neuroscience manuscript. Use this skill whenever the user asks to review, critique, or give feedback on a neuroscience paper, draft manuscript, or scientific write-up — whether they say "review my paper", "check my manuscript before submission", "act as a referee for my paper", "give me feedback on this neuroscience paper", or simply upload/paste a neuroscience manuscript and ask for comments. Also trigger when the user mentions fMRI, EEG, connectivity analysis, computational neuroscience, brain dynamics, or related topics in the context of evaluating written work. Covers all neuroscience subfields including systems, cognitive, computational, clinical, and complex-network approaches.

Configure neuroflow integrations — Miro, Google Workspace, and custom LLM providers. Use when setting up credentials, checking integration status, or guiding a user through connecting external services. Also covers e-INFRA CZ connection for Czech academic researchers.

Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.

Worker-critic agentic loop protocol — orchestrator coordinates a worker agent and a critic agent across up to 3 revision cycles to produce a vetted output for any phase.

2 hooks across 1 event