Aster v0.1.0 · macOS
Run experiments with coding agents, keep the evidence of every Run, and write the paper beside it — one project, one window.
Why Aster
A coding environment treats the repository as the unit of work. Aster treats the research project as the unit: the conversations that start the work, the Experiments and Runs that produce results, the paper that reports them, the papers it cites, the GPUs it runs on, and the deadline it is aiming for. Claude Code and Codex do the work in your terminal as usual; Aster records what they actually ran and puts the evidence next to every number they report.
Start from a Claude Code or Codex chat. An Experiment groups the results afterwards; it is not a form to fill in before you can talk to an agent.
Each Run records what actually ran, each number in the paper points to the Run behind it, and only you approve what runs next.
No Aster account or hosted service. Agents and research records stay on your computer and the hosts you connect, in plain files inside the project.
How it works
Your agent launches every experiment through a small research harness. The harness writes what it observed into .aster/ in your project, and Aster reads those records live.
Describe the experiment in a Workbench chat or the Experiment agent pane. Aster attaches the project context — Runs, metrics, open findings — as a compact chip, not a wall of text.
The agent launches training with run exec --experiment <slug> -- python train.py …. The harness runs only that command and records the launch commit, a patch of uncommitted changes, host, argv, and agent session.
Metric lines your code prints (ASTER_METRIC {…}) stream into metrics.jsonl; stdout and stderr go to train.log; the exit code decides the Run’s state.
aster research watch registers what the agent is waiting for. Aster checks local Runs every 15 s and SSH Runs every 60 s and wakes the agent when one fires, so nobody polls.
The agent records findings with evidence — a metric, log lines, a file — and proposes the next Run. The proposal appears under Needs you in the notebook; you approve or reject it.
When a number goes into paper/main.tex, the agent records a claim that ties it to a Run’s metric. Writing shows each claim as verified, stale, mismatched, or missing evidence.
Everything lives in the project as JSON, JSONL, and logs you can read, diff, and commit.
my-project/
paper/main.tex paper/references.bib
.aster/
project.json path map of the project
claims.json manuscript numbers → Run evidence
experiments/<slug>/
experiment.json question, hypothesis, metric, baseline, seeds
loop.jsonl hypothesis → Run → result → decision
proposals.jsonl next-Run proposals and your decisions
quality.jsonl findings with evidence
analysis.jsonl analyses and the Runs they used
runs/<id>/
run.json metrics.jsonl train.log code.patch
config.json media.jsonl artifacts/
library/papers/Experiments
Every Experiment has a notebook, its Runs, and an agent beside it. Agents propose; you decide what runs.
Each Experiment opens on a notebook: the plan (question, metric, baseline, seeds), Needs you (failed Runs, proposals, and findings waiting on a decision), results with mean ± spread per configuration, analyses, and your conclusion.
The Experiment agent is docked beside it. Ask it to design, analyze, explain a failure, or propose the next Run.
Live metric curves, the log, config, code patch, and media for every Run. Attach existing JSONL or CSV results, link a SLURM job, or let agents register Runs that Aster picks up on its own.
Tick 2–6 Runs and Compare lines them up: differences from the baseline at one precision per column, mismatched units or evaluation conditions, config differences, and outputs — text or image — matched by sample ID. Rate samples blind, A/B, and see how your ratings agree with the agent judge.
Give an agent an objective, a metric, and a budget of Runs and hours. It writes each hypothesis, Run, result, and decision to the loop ledger and launches only through the harness.
It stops for you before going over budget or changing the evaluation.
Writing
The paper sits beside the Runs it reports, and every number stays tied to the Run it came from.
The paper sits beside the Runs it reports: TeX source, the PDF compiled with Tectonic, and a Writing agent side by side, with SyncTeX in both directions.
Select text in the TeX source or the PDF and choose Revise. The Writing agent edits only that passage, and Compile PDF shows the result.
paper/main.tex and sent with “Lead with the result and give the numbers.” The agent reads the file and edits those lines, the source reloads, and Compile PDF shows the new abstract.The Claims drawer checks every number in the paper against the Run it cites and marks it verified, stale, mismatched, or missing evidence. It also lists result-like numbers that no Run backs.
When the paper no longer states its Run’s value, Update text fixes it in one click.
Library
Each project keeps its own papers, with a reading agent that relates them to your manuscript and code.
The project’s papers and PDFs with highlights and notes, arXiv and DOI import, and copyable citations.
Select a passage to highlight it, or choose Explain, Translate, or Ask. The Reading agent answers beside the PDF, and can summarize, critique, or relate the paper to your project.
Others
The GPUs and deadlines around your projects, the coding workspace, and the CLI your agents use.
SSH hosts, GPUs, and SLURM queues across projects, with a GPU budget per host and project: jobs over budget are submitted held and released in priority order.
Conference deadlines from the ccfddl and Hugging Face AI deadlines catalogs, in your time zone.
Chats, terminals, the editor, files, and Git for the project. Switching pages never stops an agent or terminal.
The Research | Code switch flips the sidebar to a worktree-first layout for pure coding work.
One small CLI for every experiment. The harness is a zero-dependency Node CLI that agents call to launch Runs and record claims; the app ships the same code and imports it at runtime, so the agent and the UI always agree on what a record means.
HARNESS=$(aster research harness-path)
# Launch a Run: only this command runs, and its metrics are recorded
node $HARNESS run exec --experiment lora-r16 --id seed-42 \
--metric-def loss:minimize --metric-def acc:maximize:% \
-- python train.py --lr 3e-4
# Tie a manuscript number to that Run's metric
node $HARNESS claim add --text "LoRA reaches 83.0% validation accuracy" \
--run lora-r16/seed-42 --metric acc --split val --format percent:1
# verified / stale / mismatch / missing-evidence
node $HARNESS claim auditFree and open source under the MIT License.
Open the .dmg and drag Aster into Applications.
First launch. Builds are signed ad hoc and not notarized yet, so macOS blocks the first open. Right-click Aster → Open → Open, or run:
xattr -dr com.apple.quarantine /Applications/Aster.appSign in to at least one agent CLI: Claude Code or Codex. Any other CLI agent runs in a terminal tab.
Open a project: Home → Add project and pick a folder. For a new project, create an Experiment or a manuscript and Aster writes .aster/ and paper/ for you, without overwriting existing files.
PATH to compile PDFs.Aster keeps its own data in ~/Library/Application Support/Aster and research records inside each project.
FAQ