Arbor is an autonomous research agent that proposes hypotheses, edits code, runs experiments, and retains improvements that succeed on held-out data in a growing hypothesis tree. Researchers use it to investigate problems and iteratively optimize solutions with real experiments. The catalogue skills and plugin expose Arbor's research-agent workflow to coding agents.
Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add RUC-NLPIR/Arbor --skill arbor-research-agentgit clone --depth 1 https://github.com/RUC-NLPIR/ArborWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/ruc-nlpir/arbor/arbor-research-agent)<a href="https://agentmods.dev/skills/ruc-nlpir/arbor/arbor-research-agent"><img src="https://agentmods.dev/badge/skills/ruc-nlpir/arbor/arbor-research-agent/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/ruc-nlpir/arbor/arbor-research-agent"><img src="https://agentmods.dev/badge/skills/ruc-nlpir/arbor/arbor-research-agent.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00074 | $0.01694 |
| Opus 5 | $0.00037 | $0.00847 |
| Sonnet 5 | $0.00015 | $0.00339 |
| Haiku 4.5 | $0.00007 | $0.00169 |
Grade A, and why
arbor-research-agent scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 183 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Arbor Research Agent
Use this as the single user-facing entrypoint. The user should be able to say
$arbor-research-agent plus a plain-language goal, similar to using arbor,
without knowing the internal phase skills.
This skill performs Arbor-style intake and clarification, then hands control to
arbor-agent-orchestrator.
Entry Protocol
- Treat the launch cwd as the default target project unless the user names a different path.
- Read available local context before asking: README summaries, config/eval hints, cached metrics, dataset notes, and git state.
- Decide whether the request is clear enough to start. If not, ask concise Arbor-style clarification questions before optimizing.
- Once the contract is clear, load
arbor-agent-orchestratorand continue with its phase loading order.
Do not ask for information that can be discovered safely from local files. Ask only for decisions, permissions, missing objectives, or ambiguous tradeoffs.
Intake Context Budget
Keep wrapper intake small. Its job is to determine the run contract, not to fully analyze the target project.
- Start with
pwd, git branch/status,rg --files, and concise slices such as README/config/eval metadata. - Use
rgto locate metric/eval/data hints before opening files. Prefer focusedsed -nranges over full-file reads. - Do not bulk-read long logs, notebooks, lockfiles, generated outputs, or large source files during wrapper intake.
- For training logs or progress logs, avoid raw
cat, rawgrep, or broadtail. Ifarbor-agent-toolsis available, usearbor_state.py parse-log; otherwise normalize carriage returns withtr '\r' '\n'and inspect only the metric lines needed for the contract. - Defer deep code reading to
arbor-agent-setup-intake,arbor-agent-executor, or the relevant phase skill after the orchestrator is loaded.
Clarification Gate
If any of these are missing or ambiguous after local inspection, ask before starting the optimization loop:
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 10d ago First seen · 183 lines · 74 tokens per session scan A 3405e1017ce0
arbor-research-agent is a skill published in the GitHub repository RUC-NLPIR/Arbor (1,060 stars, last pushed yesterday), licensed Apache-2.0. It adds 74 tokens to every session and 1,694 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
skill-forge
Autonome Verbesserung nach dem Autoresearch-Paradigma (Karpathy). Zwei Modi: (1) Skill-Modus — optimiert eine SKILL.md durch iterative Mutation und Evaluation. (2) Generic-Modus — optimiert beliebige Dateien gegen jede mechanische Metrik (Testabdeckung, Bundle-Size, Lighthouse-Score, Docker-Image-Größe, etc.). Zwei…
autogpt-agents
Autonomous AI agent platform for building and deploying continuous agents. Use when creating visual workflow agents, deploying persistent autonomous agents, or building complex multi-step AI automation systems.
autocontext
Iterative strategy generation and evaluation system. Use when the user wants to evaluate agent output quality, run improvement loops, queue tasks for background evaluation, check run status, inspect runtime artifacts and session branch lineage, or discover available scenarios. Provides LLM-based judging with…
autocontext-consumer
Use when an agent needs to USE knowledge Autocontext already produced - find which scenarios have knowledge, read the playbook and lessons for one, understand the on-disk file and folder layout, and move knowledge between checkouts. Host-agnostic; requires only the autoctx CLI and the filesystem.
autocontext-creator
Use when an agent needs to CREATE knowledge with Autocontext - run a scenario or plain-language task through the improvement loop, judge or improve a single output, and inspect what the run produced. Host-agnostic; requires only the autoctx CLI.
codex-autoresearch
Triage improvement work and run or resume accepted measured loops in a local project. Architecture, documentation, UX, product study, open research, taste, and one-shot fixes stay direct unless the user explicitly requests repeated measurement with a complete experiment contract.