Przejdź do treści
Biblioteka skilli
Nauka i badania · Planowanie

initializing-a-research-lab

/initializing-a-research-lab

Zakłada, dołącza lub rozbudowuje laboratorium badawcze w projekcie badacza, gdy w folderze nie ma jeszcze struktury laboratorium.

Licencja
MIT
Ostatnia zmiana repozytorium
09.11.2026

Ocena i bezpieczeństwo

100/ 100Klasa: dobry
Format25 / 25
Frontmatter z nazwą i opisem, poprawna nazwa, opis do 1024 znaków, nazwa zgodna z folderem.
Odkrywalność25 / 25
Opis mówi, kiedy użyć skilla, ma sensowną długość i jest w 3. osobie.
Treść35 / 35
Co najmniej 150 słów i do 500 linii, nagłówki, listy kroków, przykłady albo kod.
Higiena15 / 15
Bez zaślepek typu TODO, bez ścieżek Windows, z licencją.
Ryzykobrak sygnałów w skanie
Pliki w paczce
162
Skrypty
17
Komentarze HTML
1
Adresy URL w SKILL.md
0
Przestarzałość

Brak sygnałów. Repozytorium zmienione 09.11.2026, bez odwołań do wycofanych modeli i API.

Licencja

MIT, otwarta: można używać i zmieniać, z zachowaniem informacji o autorze.

Co sprawdzić przed instalacją
  • 17 skryptów w paczce. Przejrzyj je przed instalacją, bo Claude może je uruchomić.
  • SKILL.md zawiera komentarze HTML (1). Nie widać ich w podglądzie Markdown, a model je czyta.

Skrypty: scripts/init_lab.py, scripts/setup_lab_env.ps1, scripts/tests/test_init_claude_section.py, scripts/tests/test_init_columns.py, scripts/tests/test_init_connectors.py, scripts/tests/test_init_existing.py, scripts/tests/test_init_fresh.py, scripts/tests/test_init_lab_json.py, scripts/tests/test_init_migration.py, scripts/tests/test_init_output_style.py, scripts/tests/test_init_permissions.py, scripts

Co robi i kiedy

Zakłada, dołącza lub rozbudowuje laboratorium badawcze w projekcie badacza, gdy w folderze nie ma jeszcze struktury laboratorium.

Oryginalny opis (EN)

Use when the researcher wants a research lab started, joined or upgraded in this project, or asks for research work in a folder that has no research_lab/

Jak działa

  1. Overview
  2. Ask first
  3. Show, then write
  4. After writing
  5. Never, even when asked
  6. Command
Sekcje z nagłówków SKILL.md, w oryginale · 73 linii, 1 115 słów
Podgląd SKILL.md
---
name: initializing-a-research-lab
description: Use when the researcher wants a research lab started, joined or upgraded in this project, or asks for research work in a folder that has no research_lab/
---

# Initializing a Research Lab

## Overview

The setup script adds what the project lacks: folders, templates and the artifact archive, the lab section of `CLAUDE.md` (`.claude/CLAUDE.md` if it is the only one) and, on request, connectors and permission rules. It never overwrites a file; a re-run adds what is new and updates `lab.json`. A re-run also refreshes the lab section between the markers when it differs from the template, and keeps everything outside the markers.

The lab set-up has these folders under `research_lab/`: `project/`, `literature/`, `findings/`, `experiments/`, `reviews/`, `requests/`, `decisions/`, `planning/`, `automation/`, `tool_vetting/`, `templates/`, `artifacts/`, `memory/` and `.working/`. `lab.json` holds the lab's name, `roles` (the fourteen role ids, such as `research-project-manager`), `idPrefixes`, `tracks`, `preregistration` (the two commit-trailer names the registration check reads), `connectors`, `scope.restrictedFolders` and `permissions`.

A service track needs less than the whole lab. `--track idea`, `literature`, `review`, `writing` or `dissemination` (repeatable) writes only the minimal set for that track: the source table, the reviews folder and log, the artifact archive, lab memory, the templates that track uses and a full `lab.json` that lists the track in `tracks`. A re-run with another `--track`, or without one, shows its plan and adds what is new after the usual yes; `tracks` only grows, and nothing is promoted to the whole lab silently.

**Core principle:** Show the plan, then write. "Do all of it" is a yes to the lab, never to what the lab does not do.

## Ask first

Ask once; after "no questions", take and name the defaults.

1. Name, short name, description (one or two sentences), Git Large File Storage (LFS) threshold (default 10 MB).
   Also ask: "Which track: idea, literature, review, writing, dissemination, or the whole lab?" A track answer becomes `--track <name>`; "the whole lab" passes no `--track`. The researcher may name more than one track.
2. **Claude Science:** if `%USERPROFILE%\.claude-science\config.toml` exists, offer `science-ultrapowers:migrating-from-claude-science`. Never copy that folder by hand.
3. **Connectors:** tick the six http `science` entries, not the other `mcp` ones; offer `optuna` when the project brief names hyper-parameter studies. `account` connectors work after `/mcp` sign-in, `bundled` ones only in Claude Science; neither goes into `.mcp.json`.
4. Restricted folders, which no role may open (`--out-of-scope`, repeatable; the script writes them to `scope.restrictedFolders` in `lab.json`).
5. Whether to set up the figure and poster environment.
6. **Permission rules** (default no): `--permissions` writes the lab's literature-site `WebFetch` rules and any `--allow-domain` (offer those a migration's `PERMISSIONS.md` approved) to `.claude/settings.json` only when absent; else it prints a JSON snippet to merge by hand.

## Show, then write

1. Run the script without `--yes` and show its plan word for word.
2. Run it with `--yes` after a yes; an earlier go-ahead covers a plan of only what was asked for, and refusing part never blocks the rest.
3. In a folder that is not a git repository, offer `--git-init`.
4. Show the environment script's `-DryRun` output; the researcher runs it (it asks before each change).

### The two column additions

On every lab, whatever its layout, the plan can also list two column additions, shown as `header` and `rewrite` lines: the `Basis` column on `findings/FINDINGS_LOG.md` (`header <file>: (none) -> Basis`, and `rewrite <file>: Basis, <n> rows: (empty) -> exploratory` for the rows already there) and the `Input checksums` column on `experiments/EXPERIMENT_REGISTER.md` (`header <file>: (none) -> Input checksums`; existing rows stay empty there). A table that lacks its anchor column (`Type` or `Environment`) is left as it is and the plan carries a warning. A lab that already has both columns plans nothing. No other record is edited.

### A lab in the 0.1 layout

Init migrates a lab in the 0.1 layout (a numbered folder such as `02_findings_ledger/` under `research_lab/`). <!-- legacy -->
The plan lists each change as a `rename`, `header`, `rewrite` or `refresh` line, and init applies it only after the researcher confirms with `--yes`. The migration renames folders and files with `git mv` when the lab is tracked. It rewrites table headers, status tokens, the memory keys `basis` and `type`, and `lab.json`. It never edits a free-text cell or an existing identifier. A record kept in sections (rows, a note, more rows) is one table and is migrated to its last row. A cell that starts with a 0.1 status word and holds more text is left as it is and reported, so the researcher edits it by hand. The archive's records (`artifacts/versions.jsonl`, `artifacts/artifacts.jsonl`) follow the renames, so a linked file keeps its link. A project that tracks its `references/` folder in git keeps it tracked: the ignore line for it is never added, and the plan says so. A template, a folder guide or the text above a record's table that is still exactly as 0.1 wrote it is refreshed to the current text; a file the researcher edited is kept, with a warning. It stops for a folder when the old and the new name both exist, and reports `unexpected layout`. Until the migration runs, every lab script exits with the message `this lab uses the 0.1 layout; run science-ultrapowers:initializing-a-research-lab to migrate it`.

### The scholarly output style

Init installs the scholarly output style at `.claude/output-styles/scholarly-explanatory.md` when the file is absent. It prints the line `"outputStyle": "Scholarly Explanatory"` for the researcher to add to `.claude/settings.json`. It writes nothing when the file exists, and says so.

## After writing

Tell the researcher to:
- fill in `research_lab/project/PROJECT_BRIEF.md`;
- commit `research_lab/`, the new root files and each file the report says to review (in an existing repository init commits nothing);
- approve `.mcp.json` when asked, then sign in via `/mcp`;
- remove the listed `.claude` duplicates by hand after checking them for local changes;
- install `git lfs` if the report says so;
- add the printed `"outputStyle"` line to `.claude/settings.json` to write in the scholarly style.

## Never, even when asked

| Request | What you do |
|---|---|
| "Put the key in a config file" | No secret in a file or on a command line. The Jev key lives in the `TYPESAFE_API_KEY` environment variable, set by the researcher or the environment script's hidden input. Rotate a key pasted into chat. |
| "Clean out the old copies" | You move and delete nothing under `.claude/agents` or `.claude/skills`. List them. |
| "Allow running python" | No permission rule for commands. |
| "Every connector" | The `mcp` entries at most. |

Never edit `~/.claude`, an existing `.claude/settings.json`, or any file the plan marks `skip`.

## Command

```bash
python "${CLAUDE_PLUGIN_ROOT}/skills/initializing-a-research-lab/scripts/init_lab.py" [--project-root <dir>] --name "<name>" --short-name <short> --description "<one or two sentences>" [--connectors science,<id>,...] [--out-of-scope <folder> ...] [--track <idea|literature|review|writing|dissemination> ...] [--lfs-threshold-mb 10] [--git-init] [--permissions [--allow-domain <host> ...]] [--yes] [--json]
pwsh -File "${CLAUDE_PLUGIN_ROOT}/skills/initializing-a-research-lab/scripts/setup_lab_env.ps1" -ProjectRoot <project> -DryRun
```
Pliki w paczce (162)
  • SKILL.md 8 KB
  • connectors.json 17 KB
  • scripts/init_lab.py 100 KB
  • scripts/setup_lab_env.ps1 9 KB
  • scripts/tests/test_init_claude_section.py 7 KB
  • scripts/tests/test_init_columns.py 9 KB
  • scripts/tests/test_init_connectors.py 10 KB
  • scripts/tests/test_init_existing.py 13 KB
  • scripts/tests/test_init_fresh.py 22 KB
  • scripts/tests/test_init_lab_json.py 7 KB
  • scripts/tests/test_init_migration.py 28 KB
  • scripts/tests/test_init_output_style.py 3 KB
  • scripts/tests/test_init_permissions.py 9 KB
  • scripts/tests/test_init_references.py 2 KB
  • scripts/tests/test_init_refresh.py 15 KB
  • scripts/tests/test_init_scope.py 3 KB
  • scripts/tests/test_init_track.py 7 KB
  • scripts/tests/test_init_track_texts.py 3 KB
  • scripts/tests/test_setup_lab_env.py 3 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/README.md 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/lab.json 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/.scratch/.gitkeep 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/00_project_knowledge/PROJECT_BRIEF.md 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/00_project_knowledge/README.md 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/01_reference_registry/EXCEPTIONS.md 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/01_reference_registry/README.md 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/01_reference_registry/REGISTRY.csv 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/01_reference_registry/audits/.gitkeep 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/01_reference_registry/evidence_cards/EC-0001.md 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/02_findings_ledger/CANONICAL_NUMBERS.md 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/02_findings_ledger/FINDINGS.md 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/02_findings_ledger/LAB_NOTEBOOK.md 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/02_findings_ledger/README.md 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/03_experiments/EXPERIMENT_REGISTER.md 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/03_experiments/README.md 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/03_experiments/protocols/.gitkeep 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/04_reviews/README.md 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/04_reviews/REVIEW_LOG.md 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/04_reviews/reports/.gitkeep 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/05_access_requests/ACCESS_REQUESTS.md 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/05_access_requests/README.md 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/06_decisions/DECISIONS.md 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/06_decisions/README.md 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/06_decisions/lab_meetings/2026-01-05_kickoff.md 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/07_project_management/README.md 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/07_project_management/RISK_REGISTER.md 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/07_project_management/ROADMAP_STATUS.md 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/07_project_management/TASK_BOARD.md 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/08_decision_log/JEV_CALLS.jsonl 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/08_decision_log/README.md 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/08_decision_log/thresholds.json 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/09_skill_vetting/README.md 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/artifacts/.gitattributes 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/artifacts/artifacts.jsonl 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/artifacts/versions.jsonl 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/artifacts/store/artifacts.jsonl 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/artifacts/store/dependencies.jsonl 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/artifacts/store/verification_checks.jsonl 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/artifacts/store/versions.jsonl 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/memory/MEMORY.md 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/memory/README.md 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/memory/artifact/.gitkeep 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/memory/project/a-pitfall.md 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/memory/project/another-pitfall.md 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/migration/WHERE_WE_LEFT_OFF.md 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/templates/EVIDENCE_CARD.md 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/templates/H_CARD.md 1 KB
  • scripts/tests/fixtures/lab_0_1/research_lab/templates/SESSION_KICKOFF.md 1 KB
  • scripts/tests/fixtures/templates_0_1/research_lab/README.md 2 KB
  • scripts/tests/fixtures/templates_0_1/research_lab/00_project_knowledge/PROJECT_BRIEF.md 1 KB
  • scripts/tests/fixtures/templates_0_1/research_lab/00_project_knowledge/README.md 1 KB
  • scripts/tests/fixtures/templates_0_1/research_lab/01_reference_registry/EXCEPTIONS.md 1 KB
  • scripts/tests/fixtures/templates_0_1/research_lab/01_reference_registry/README.md 1 KB
  • scripts/tests/fixtures/templates_0_1/research_lab/01_reference_registry/REGISTRY.csv 1 KB
  • scripts/tests/fixtures/templates_0_1/research_lab/02_findings_ledger/CANONICAL_NUMBERS.md 1 KB
  • scripts/tests/fixtures/templates_0_1/research_lab/02_findings_ledger/FINDINGS.md 1 KB
  • scripts/tests/fixtures/templates_0_1/research_lab/02_findings_ledger/LAB_NOTEBOOK.md 1 KB
  • scripts/tests/fixtures/templates_0_1/research_lab/02_findings_ledger/README.md 1 KB
  • scripts/tests/fixtures/templates_0_1/research_lab/03_experiments/EXPERIMENT_REGISTER.md 1 KB
  • scripts/tests/fixtures/templates_0_1/research_lab/03_experiments/README.md 1 KB
  • scripts/tests/fixtures/templates_0_1/research_lab/04_reviews/README.md 1 KB
  • scripts/tests/fixtures/templates_0_1/research_lab/04_reviews/REVIEW_LOG.md 1 KB
  • scripts/tests/fixtures/templates_0_1/research_lab/05_access_requests/ACCESS_REQUESTS.md 1 KB
  • scripts/tests/fixtures/templates_0_1/research_lab/05_access_requests/README.md 1 KB
  • scripts/tests/fixtures/templates_0_1/research_lab/06_decisions/DECISIONS.md 1 KB
  • scripts/tests/fixtures/templates_0_1/research_lab/06_decisions/README.md 1 KB
  • scripts/tests/fixtures/templates_0_1/research_lab/07_project_management/README.md 1 KB
  • scripts/tests/fixtures/templates_0_1/research_lab/07_project_management/RISK_REGISTER.md 1 KB
  • scripts/tests/fixtures/templates_0_1/research_lab/07_project_management/ROADMAP_STATUS.md 1 KB
  • scripts/tests/fixtures/templates_0_1/research_lab/07_project_management/TASK_BOARD.md 1 KB
  • scripts/tests/fixtures/templates_0_1/research_lab/08_decision_log/JEV_CALLS.jsonl 1 KB
  • scripts/tests/fixtures/templates_0_1/research_lab/08_decision_log/README.md 1 KB
  • scripts/tests/fixtures/templates_0_1/research_lab/08_decision_log/thresholds.json 1 KB
  • scripts/tests/fixtures/templates_0_1/research_lab/09_skill_vetting/README.md 1 KB
  • scripts/tests/fixtures/templates_0_1/research_lab/memory/MEMORY.md 1 KB
  • scripts/tests/fixtures/templates_0_1/research_lab/memory/README.md 3 KB
  • scripts/tests/fixtures/templates_0_1/research_lab/templates/CITATION_AUDIT.md 1 KB
  • scripts/tests/fixtures/templates_0_1/research_lab/templates/DEVILS_ADVOCATE_TEMPLATE.md 1 KB
  • scripts/tests/fixtures/templates_0_1/research_lab/templates/EVIDENCE_CARD.md 1 KB
  • scripts/tests/fixtures/templates_0_1/research_lab/templates/HANDOFF.md 1 KB
  • scripts/tests/fixtures/templates_0_1/research_lab/templates/H_CARD.md 1 KB
  • scripts/tests/fixtures/templates_0_1/research_lab/templates/LAB_MEETING.md 1 KB
  • scripts/tests/fixtures/templates_0_1/research_lab/templates/METHOD_PROPOSAL.md 1 KB
  • scripts/tests/fixtures/templates_0_1/research_lab/templates/PREREGISTRATION.md 1 KB
  • scripts/tests/fixtures/templates_0_1/research_lab/templates/REFEREE_REPORT.md 1 KB
  • scripts/tests/fixtures/templates_0_1/research_lab/templates/REVIEW_ACADEMIC.md 1 KB
  • scripts/tests/fixtures/templates_0_1/research_lab/templates/SESSION_KICKOFF.md 1 KB
  • scripts/tests/fixtures/templates_0_1/research_lab/templates/SESSION_STANDUP.md 1 KB
  • templates/CLAUDE_SECTION.md 3 KB
  • templates/gitignore_lines.txt 1 KB
  • templates/lab.json 1 KB
  • templates/permissions.json 1 KB
  • templates/research_lab/README.md 2 KB
  • templates/research_lab/.working/.gitkeep 1 KB
  • templates/research_lab/artifacts/archive/.gitkeep 1 KB
  • templates/research_lab/automation/JEV_CALLS.jsonl 1 KB
  • templates/research_lab/automation/README.md 1 KB
  • templates/research_lab/automation/thresholds.json 1 KB
  • templates/research_lab/decisions/DECISION_LOG.md 1 KB
  • templates/research_lab/decisions/README.md 1 KB
  • templates/research_lab/decisions/lab_meetings/.gitkeep 1 KB
  • templates/research_lab/experiments/EXPERIMENT_REGISTER.md 1 KB
  • templates/research_lab/experiments/README.md 1 KB
  • templates/research_lab/experiments/feasibility/.gitkeep 1 KB
  • templates/research_lab/experiments/preregistrations/.gitkeep 1 KB
  • templates/research_lab/findings/FINDINGS_LOG.md 1 KB
  • templates/research_lab/findings/KEY_RESULTS.md 1 KB
  • templates/research_lab/findings/LAB_NOTEBOOK.md 1 KB
  • templates/research_lab/findings/README.md 1 KB
  • templates/research_lab/literature/EXCEPTIONS.md 1 KB
  • templates/research_lab/literature/README.md 1 KB
  • templates/research_lab/literature/SOURCES.csv 1 KB
  • templates/research_lab/literature/citation_audits/.gitkeep 1 KB
  • templates/research_lab/literature/evidence_records/.gitkeep 1 KB
  • templates/research_lab/literature/search_logs/.gitkeep 1 KB
  • templates/research_lab/memory/artifact/.gitkeep 1 KB
  • templates/research_lab/memory/project/.gitkeep 1 KB
  • templates/research_lab/planning/README.md 1 KB
  • templates/research_lab/planning/RISK_REGISTER.md 1 KB
  • templates/research_lab/planning/ROADMAP_STATUS.md 1 KB
  • templates/research_lab/planning/WORK_PLAN.md 1 KB
  • templates/research_lab/project/PROJECT_BRIEF.md 1 KB
  • templates/research_lab/project/README.md 1 KB
  • templates/research_lab/requests/ACCESS_REQUESTS.md 1 KB
  • templates/research_lab/requests/README.md 1 KB
  • templates/research_lab/reviews/README.md 1 KB
  • templates/research_lab/reviews/REVIEW_LOG.md 1 KB
  • templates/research_lab/reviews/reports/.gitkeep 1 KB
  • templates/research_lab/templates/ADVERSARIAL_REVIEW.md 2 KB
  • templates/research_lab/templates/ARCHIVE_MANIFEST.md 1 KB
  • templates/research_lab/templates/CITATION_AUDIT.md 1 KB
  • templates/research_lab/templates/EVIDENCE_RECORD.md 1 KB
  • templates/research_lab/templates/HYPOTHESIS_RECORD.md 1 KB
  • templates/research_lab/templates/INTERNAL_REVIEW.md 1 KB
  • templates/research_lab/templates/METHOD_PROPOSAL.md 1 KB
  • templates/research_lab/templates/PREREGISTRATION.md 2 KB
  • templates/research_lab/templates/PROTOCOL_ADHERENCE_REVIEW.md 1 KB
  • templates/research_lab/templates/REFEREE_REPORT.md 1 KB
  • templates/research_lab/templates/RESPONSE_MATRIX.md 1 KB
  • templates/research_lab/templates/SYNTHESIS_MATRIX.md 1 KB
  • templates/research_lab/templates/TASK_REPORT.md 2 KB
  • templates/research_lab/tool_vetting/README.md 1 KB

Wersje

W bazie nie ma innych skilli o nazwie initializing-a-research-lab.

Migawka bazy z 13.11.2026 · wersja paczki 280959f0099b (skrót SHA-256 całej paczki)

Pobrania w ostatnich 30 dniach: 0 · ranking

Zgłoś problem z tym skillem: szkodliwa treść, błędny opis albo prośba autora o usunięcie.

Podobne skille

  • 100running-the-lab
    Obsługuje zwroty badacza do laboratorium: role, etapy, rekordy i zadania badawcze, zanim cokolwiek zostanie przydzielone lub zapisane.
  • 100presenting-research
    Przygotowuje slajdy, plakat, infografikę lub schemat architektury z zapisów laboratorium albo z artykułu.
  • 100materio-workflow-planner
    Dostosowuje plan badań z badaczem, ustala kolejność zadań i założenia, a potem zapisuje WORKFLOW.md sprawdzony względem laboratorium.
  • 86lab-start
    Wprowadza nowego właściciela w laboratorium byolab i prowadzi go do pierwszego ocenionego komponentu.
  • 100plan-research-review
    Porównuje plan badawczy z istniejącym kodem i pyta badacza o każdą lukę, gdy o to poprosi lub zgodnie z instrukcją sondy.
  • 90agent-research-lab
    Prowadzi laboratorium agentów badaczy, którzy równolegle szukają, niezależnie weryfikują źródła i syntetyzują odpowiedź z cytatami, gdy każde kluczowe twierdzenie musi prowadzić do dowodu.

Uwagi i zgłoszenia

Wiadomość trafi do Bartosza Sokolińskiego. Adres e-mail jest potrzebny tylko do odpowiedzi. Szczegóły w polityce prywatności.